Quartz 4

❯

Folder: notes

832 items under this folder.

Sep 25, 2026
Antibody-specific fine-tuning improves framework preservation and hotspot targeting in RFdiffusion
- design/antibodies
- training/fine-tuning
Sep 25, 2026
De novo designed binders can have unbound ensembles that differ from the bound state
Sep 25, 2026
General-purpose language models underperform specialist models on tested epitope identification tasks
- antibodies/recognition
- evidence/generalization
Sep 25, 2026
Linear models remain competitive on epistasis-enriched protein fitness tests
- prediction/variant-effects
- evidence/generalization
Sep 25, 2026
Surface-complementary seeds expand the structural diversity of diffusion-designed protein binders
Sep 07, 2026
Cysteines are the most common conserved residue in E. coli
- evolution/selection
Sep 07, 2026
Disentangled attention for learning joint sequence-structure representations
- model-design/multimodal
- model-design/architectures
Sep 07, 2026
ESM-IF, but not other inverse folding models, has learned some evolutionary constraints from sequence databases
Sep 07, 2026
ESMFold has massive activations that disproportionately contribute to output quality
- model-analysis/interpretability
- prediction/structure
Sep 07, 2026
ESMFold pLDDT weakly correlates with intra-family differences in stability and experimental success, but not differences in cooperativity
Sep 07, 2026
Feynman-Kac steering
- inference/guidance
- model-design/generative-models
Sep 07, 2026
Fine-tuning PLMs on rank is more effective than fine-tuning on property (regression)
Sep 07, 2026
Hard constraints can be enforced during diffusion by applying to fully denoised predictions and renoising
- inference/guidance
- design/backbones
Sep 07, 2026
Free antibody light chains can homodimerize
- biophysics/interactions
Sep 04, 2026
Kinked CDRH3 loops
- antibodies/architecture
- antibodies/nanobodies
Aug 25, 2026
Antibody thermostability can decrease during affinity maturation, up to a threshold
- antibodies/maturation
- biophysics/stability
Aug 25, 2026
A small number of high-energy hot spot interactions can dominate protein-protein binding energy
- biophysics/interactions
Aug 25, 2026
Ab-Ag inverse folding methods benefit from pretraining
Aug 25, 2026
Accounting for the unbound state improves inverse-folding prediction of binding-affinity changes
- prediction/binding
- design/inverse-folding
Aug 25, 2026
AlphaFold3 ipTM can distinguish between antibody binders and nonbinders
Aug 25, 2026
Antibodies often target cognate ligand epitopes without reproducing ligand contacts
- antibodies/recognition
- biophysics/interactions
Aug 25, 2026
Antibody affinity can sometimes be improved by stabilization
- design/antibodies
- biophysics/stability
Aug 25, 2026
Ensembling structure prediction methods for filtering improves recovery
Aug 25, 2026
Fine-tuning almost always improves property prediction
- training/fine-tuning
- inference/feature-extraction
Aug 25, 2026
Fine-tuning leads to small changes in parameter values in L2 space
- training/fine-tuning
Aug 25, 2026
Low-rank adaptation causes models to learn less and forget less
- training/fine-tuning
Aug 25, 2026
Masked LMs can be fine-tuned starting from autoregressive LMs, but not vice-versa
- training/fine-tuning
- model-design/architectures
Aug 25, 2026
Most assignments of genetic novelty due to homolog detection failure
- evolution/homology
- evolution/selection
Aug 25, 2026
PLM searches can improve remote-homology detection beyond profile HMMs
- evolution/homology
- inference/feature-extraction
Aug 25, 2026
Prefusion stabilization often requires multiple mutations, but single substitutions can suffice
- design/sequence-generation
- biophysics/stability
Aug 25, 2026
Pretraining performance does not capture effectiveness on downstream tasks
Aug 25, 2026
Structural tokens improve zero-shot variant effect prediction in ESM3, but only when structures are not computationally derived
Aug 25, 2026
The effect of including both sequence and structural features can be synergistic
- model-design/multimodal
- design/inverse-folding
Aug 25, 2026
Zero-shot protein fitness prediction using sequence-only NNs can be improved by averaging predictions from many orthologs
- prediction/variant-effects
- inference/ensembling
Aug 25, 2026
Zero-shot protein stability prediction using inverse folding models can be improved by subtracting predictions from residue in isolation
- prediction/stability-expression
- design/inverse-folding
Aug 25, 2026
Antibody hydrophobicity sometimes correlates with polyreactivity
- design/developability
- antibodies/recognition
Aug 25, 2026
Antibody thermostability is not correlated with hydrophobicity
- biophysics/stability
- design/developability
Aug 25, 2026
Antibody thermostability is not correlated with polyreactivity
- biophysics/stability
- design/developability
Aug 25, 2026
Somatic hypermutation correlates with lower hydrophobicity
- antibodies/maturation
- design/developability
Aug 25, 2026
Somatic hypermutation correlates with lower polyreactivity
Aug 25, 2026
Annealing epitope guidance reduces steric clashes in de novo antibody design
- design/antibodies
- inference/guidance
Aug 25, 2026
Antigen size biases AlphaFold3 antibody-antigen confidence
- prediction/confidence
- prediction/complexes
Aug 25, 2026
Protein backbone design methods show fold-level novelty by recycling memorized subdomain structures
- design/backbones
- evidence/generalization
Aug 25, 2026
Structural datasets overrepresent a small number of epitopes
- evidence/datasets
- antibodies/nanobodies
Aug 25, 2026
Boltz-2 intermediate representations can predict protein-protein binding affinity
- prediction/binding
- inference/feature-extraction
Aug 25, 2026
CDRH3 is necessary but insufficient to guarantee binding to a particular epitope
- antibodies/recognition
- antibodies/repertoires
Aug 25, 2026
Computational antibody design is more reliable when HCDR3 is held fixed
- design/antibodies
- evidence/design-validation
Aug 25, 2026
Computational ranking of HCDR3 clones can underperform random selection
- evidence/design-validation
- design/antibodies
Aug 25, 2026
De novo antibody design
- design/antibodies
Aug 25, 2026
Ligand-mimicry filtering improves epitope-targeted antibody discovery
- design/antibodies
- evidence/design-validation
Aug 25, 2026
NGS sequence abundance does not correlate with binding affinity
- evidence/measurements
- design/directed-evolution
Aug 25, 2026
The Boltz-2 affinity module cannot be effectively repurposed for PPI affinity prediction
- prediction/binding
- evidence/generalization
Aug 25, 2026
ipTM is sensitive to construct length even when predicted interfaces are unchanged
- prediction/confidence
- prediction/complexes
Aug 19, 2026
Binding affinity of protein-protein interactions correlates with buried surface area
- biophysics/interactions
Aug 19, 2026
Clonal expansion correlates with antigen binding
Aug 19, 2026
Fab
- antibodies/architecture
Aug 19, 2026
Long CDRH3 loops are enriched in some polyreactive repertoires but are not strongly predictive alone
Aug 19, 2026
Most nanobodies do not bind using all three CDRs
- antibodies/nanobodies
- antibodies/recognition
Aug 19, 2026
Conventional NMR structure bundles are not thermodynamic ensembles
- evidence/measurements
- biophysics/ensemble-analysis
Aug 19, 2026
Allostery can be mediated by changes in thermal fluctuations without a mean structural change
- biophysics/dynamics
Aug 19, 2026
Fine-tuned protein language models can struggle to generalize to positions absent from fine-tuning data
Aug 19, 2026
Naive concatenation can underperform learned sequence-structure fusion
- model-design/multimodal
- inference/feature-extraction
Aug 19, 2026
Pretraining contributes nearly nothing to performance when fine-tuning protein language models in data-rich situations
- training/fine-tuning
- evidence/datasets
Aug 19, 2026
Unconditional design does not consistently improve with model scale
- design/sequence-generation
- training/pretraining-and-scaling
Aug 19, 2026
Raw folding-model confidence is not a general protein stability predictor
- prediction/confidence
- prediction/stability-expression
Aug 19, 2026
Confidence metrics are not general binding-affinity predictors
Aug 19, 2026
ROC-AUC is prevalence-invariant but does not measure deployment precision
- evidence/measurements
Aug 19, 2026
SaProt
- model-design/multimodal
- prediction/variant-effects
Aug 19, 2026
Structure prediction uncertainty metrics as energy functions
- prediction/confidence
- model-design/generative-models
Aug 19, 2026
Training inverse folding and diffusion models exclusively on predicted protein structures worsens performance due to how locally perfect they are
Jul 28, 2026
AF3 binding affinity predictions are orthogonal to those made by force fields and other neural networks
- prediction/binding
- inference/ensembling
Jul 28, 2026
All-atom structure and affinity predictors partially generalize to point mutants
Jul 28, 2026
Antibody LMs are worse for expression prediction than generic PLMs
Jul 28, 2026
Coevolutionary patterns in multiple sequence alignments do not contribute to protein-protein complex prediction
- prediction/complexes
- inference/conditioning
Jul 28, 2026
Directed evolution can uncover beneficial sequence modifications that might be missed by inverse folding
- design/directed-evolution
- design/inverse-folding
Jul 28, 2026
Directed evolution
- design/directed-evolution
Jul 28, 2026
Inverse folding models favor archaeal proteins while sequence-only PLMs favor eukaryotic proteins
- evidence/datasets
- design/inverse-folding
Jul 28, 2026
Inverse folding selects more evolutionarily conserved design positions than sequence-only protein language models
Jul 28, 2026
Inverse-folding-guided site-saturation libraries outperform random mutagenesis
Jul 28, 2026
Large-scale measurements, such as Kd and IC50, from different assays correlate weakly in different studies
- evidence/measurements
- prediction/binding
Jul 28, 2026
MSA Transformer
- training/pretraining-and-scaling
- model-design/architectures
Jul 28, 2026
MSA protein language models can identify intermolecular contacts
- prediction/structure
- prediction/complexes
Jul 28, 2026
PLMs are biased by uneven distribution of sequence data in datasets such as UniRef and UniProt
- evidence/datasets
- training/pretraining-and-scaling
Jul 28, 2026
PLMs learn family-specific protein contacts from sequence context windows of about 20-40 amino acids
- model-analysis/interpretability
- evidence/generalization
Jul 28, 2026
Protein backbone design diffusion models can be repurposed for fitting structures into electron density
Jul 28, 2026
Protein language models learn structure-level features, including disorder, in later layers
- model-analysis/interpretability
- model-analysis/representation-geometry
Jul 28, 2026
Query-conditioned outer product
- model-design/architectures
- prediction/structure
Jul 28, 2026
Random splits overestimate protein language model generalization on antibody expression prediction
- evidence/generalization
- prediction/stability-expression
Jul 28, 2026
Rubisco
- design/enzymes
- biophysics/catalysis
Jul 28, 2026
Sequence-only protein language models implicitly cluster proteins at fineness levels that increase with size
- model-analysis/interpretability
- training/pretraining-and-scaling
Jul 28, 2026
Structure prediction and design tools trained on monomers generalize to oligomers
Jul 28, 2026
Thermostability prediction from protein language model representations peaks in the last few layers
- prediction/stability-expression
- inference/feature-extraction
Jul 28, 2026
Training antibody language models on normalized mutation frequencies improves zero-shot expression prediction
Jul 28, 2026
Zero-shot performance of PLMs, but not inverse folding models, correlates with number of homologs available for training
- prediction/variant-effects
- evidence/datasets
Jul 22, 2026
Distograms can be used as loss functions for protein design by hallucination
- design/sequence-generation
- inference/guidance
Jul 22, 2026
Protein property prediction using PLMs does not benefit from scale except when predicting inferring features of either structural or sparsely populated sequence families
- training/pretraining-and-scaling
- prediction/variant-effects
Jul 22, 2026
Quanta hypothesis
- training/pretraining-and-scaling
- model-analysis/interpretability
Jul 22, 2026
Effective sample size
- biophysics/thermodynamics
- inference/guidance
Jul 20, 2026
AF3-generation methods incorporate MSA information exclusively into the pair representation
- model-analysis/interpretability
- prediction/structure
Jul 20, 2026
AlphaFold3 inference benefits more from phylogenetically diverse MSAs than deep redundant MSAs
Jul 20, 2026
Different immune repertoires converge on the same V genes for specific antigens
- antibodies/repertoires
Jul 20, 2026
Experimentally successful miniprotein binders have higher predicted stability
Jul 20, 2026
Explicit germline encoding improves prediction of non-germline CDRH3 residues
- training/pretraining-and-scaling
- antibodies/repertoires
Jul 20, 2026
Explicit germline encoding improves zero-shot antibody-affinity prediction
- prediction/binding
- antibodies/repertoires
Jul 20, 2026
Heavy-chain templates and self-distillation improve nanobody structure prediction
Jul 20, 2026
MSA-conditioned AlphaFold3 representations compress as the Pairformer proceeds
- model-analysis/representation-geometry
- prediction/structure
Jul 20, 2026
MSAs make AlphaFold3 predictions robust to extensive mutation and deletion
Jul 20, 2026
Most computational protein-stability metrics correlate poorly with experimental dG
- prediction/stability-expression
- evidence/design-validation
Jul 20, 2026
PLM pseudoperplexity indicates stability-prediction accuracy
- prediction/confidence
- prediction/stability-expression
Jul 20, 2026
Protein language models are in-context learners that default to retrieval over reasoning
- model-analysis/interpretability
- evidence/generalization
Jul 20, 2026
Removing MSAs collapses AlphaFold3's latent space
- model-analysis/representation-geometry
- prediction/structure
Jul 20, 2026
Sparse autoencoders recover protein-family and gene-ontology features from PLM representations
- model-analysis/interpretability
Jul 20, 2026
Stronger diffusion guidance reduces diversity of generated outputs
- inference/guidance
- design/sequence-generation
Jul 20, 2026
Uncrowded hypervolume improvement
- design/directed-evolution
- inference/sampling-and-search
Jul 19, 2026
Inversion of protein folding neural networks
- inference/guidance
- design/sequence-generation
Jul 19, 2026
Protein sequence and structure compression using VQ-VAE tokens is prone to codebook collapse
- model-design/multimodal
- model-design/generative-models
Jul 19, 2026
Straight-through estimator
Jul 19, 2026
Variational autoencoders
- model-design/generative-models
- model-design/multimodal
Jul 17, 2026
Aggregate benchmark correlations can mask weak within-category performance
- evidence/measurements
- prediction/variant-effects
Jul 17, 2026
Antibody-antigen complex prediction scales with training data and compute
Jul 17, 2026
Antibody-antigen modeling by diffusion-based structure prediction is data-limited
- prediction/complexes
- evidence/datasets
Jul 17, 2026
Cell-free systems have higher expression rates for minibinders than cell-based systems
Jul 17, 2026
Conformational entropy in antibodies decreases during affinity maturation
- antibodies/maturation
- biophysics/dynamics
Jul 17, 2026
Different AlphaFold3 clones have differently calibrated confidence heads
- prediction/confidence
- prediction/complexes
Jul 17, 2026
Distinct fitness measurements of the same variants of the same protein vary wildly in Spearman
- evidence/measurements
- prediction/variant-effects
Jul 17, 2026
Effectiveness of filtering metrics for de novo minibinder design vary by target
- evidence/design-validation
- design/binders
Jul 17, 2026
Gating template use during antibody CDR design increases success rates
- design/antibodies
Jul 17, 2026
Interface contact score
- prediction/confidence
- prediction/complexes
Jul 17, 2026
Language models cannot extrapolate to functional novelty or ultra-high-fitness variants
Jul 17, 2026
Linear probes trained on SAE features from intermediate steps of protein structure prediction neural networks can be used for property prediction
- inference/feature-extraction
- model-analysis/interpretability
Jul 17, 2026
Protein language models have an implicit understanding of the uncertainty of their predictions
- prediction/confidence
Jul 17, 2026
Selectivity, specificity, and conformational constraints can be encoded in loss functions during protein design
- inference/guidance
- design/sequence-generation
Jul 17, 2026
Antibody structure prediction improved with AlphaFold2 features
- prediction/antibody-structure
- model-design/multimodal
Jun 23, 2026
Predicted aligned error
- prediction/confidence
Jun 23, 2026
TM-score
- evidence/measurements
- prediction/confidence
Jun 12, 2026
Joint sequence-structure diffusion or flow matching models show superior performance when designing structure at the beginning and sequence near the end
May 10, 2026
Model type less important than data quantity when predicting catalytic properties of protein sequences
- prediction/function
- evidence/datasets
May 06, 2026
Buried CDR residue mutations have similar effects on expression in other globular proteins
- design/developability
- evolution/mutation-effects
May 06, 2026
CDR sequence changes are enough to destabilize antibodies
- biophysics/stability
- antibodies/architecture
May 06, 2026
CDR1 and CDR2 contribute more to antibody stability than CDR3
- biophysics/stability
- antibodies/architecture
May 06, 2026
ESM-IF outperforms ProteinMPNN on Ab inverse folding and affinity prediction
May 06, 2026
Fold stability is more strongly correlated with a protein's ability to refold after denaturing than its melting temperature
- biophysics/stability
- biophysics/folding
May 06, 2026
Inverse folding model logits can zero-shot predict expression changes arising from in antibody CDRs in cases where ground truth structure is known
- prediction/stability-expression
- design/inverse-folding
May 06, 2026
iPAE is anticorrelated with number of interface H-bonds
- prediction/confidence
- biophysics/interactions
Apr 28, 2026
Antibody language model log-likelihood is a decent surrogate for sequence reasonableness
- prediction/confidence
Apr 28, 2026
Aromatic and polar residues make up disproportionate number of antigen epitopes
- antibodies/recognition
- biophysics/interactions
Apr 28, 2026
Dynamics and specificity are inversely related
- biophysics/dynamics
- biophysics/interactions
Apr 28, 2026
Individual antibody paratopes can recognize and bind to multiple epitopes
- antibodies/recognition
- antibodies/nanobodies
Apr 28, 2026
Subset of antigen residues contribute disproportionately to antibody binding affinity
- antibodies/recognition
- biophysics/interactions
Apr 28, 2026
Subset of mutations obtained during affinity maturation contribute disproportionately to affinity
- antibodies/maturation
- antibodies/recognition
Apr 21, 2026
Affinity maturation causes antibody-antigen interfaces to be more similar to generic PPI interfaces
Apr 21, 2026
Affinity matured antibodies can adopt distinct binding poses from germline precursors
- antibodies/maturation
- antibodies/recognition
Apr 21, 2026
Affinity-specificity tradeoff in antibodies is partially mediated by varying the magnitude of rigidification
Apr 21, 2026
Agonist antibodies benefit from low affinity due to antigen clustering
- cell-biology/therapeutic-mechanisms
- biophysics/interactions
Apr 21, 2026
AlphaFold2 recapitulates interaction biases from PDB when modeling antibodies
- evidence/generalization
- prediction/complexes
Apr 21, 2026
Antibodies and BCRs have different binding kinetics
- biophysics/interactions
- cell-biology/immune-signaling
Apr 21, 2026
Antibodies from different repertoires within a given species targeting the same epitope often have the same light chain isotype and share critical residues
Apr 21, 2026
Antibody on and off rates are capped by the affinity maturation process
- antibodies/maturation
- biophysics/interactions
Apr 21, 2026
Antibody-antigen binding modes are not necessarily defined by induced fit
- antibodies/recognition
- biophysics/dynamics
Apr 21, 2026
Antigen context improves CDRH3 structure prediction
Apr 21, 2026
CDR3 far more likely to be part of paratope of nanobodies compared to antibodies
- antibodies/nanobodies
- antibodies/recognition
Apr 21, 2026
CDR3 loops encode specificity while CDR1 and CDR2 encode cross-reactivity
- antibodies/recognition
- antibodies/architecture
Apr 21, 2026
CDRH3 conformational rigidity reduces cross reactivity
Apr 21, 2026
CDRH3 has most contacts with antigen
- antibodies/recognition
- antibodies/architecture
Apr 21, 2026
Changes to hydrophobicity and charge correlate with antibody escape
- prediction/variant-effects
- antibodies/recognition
Apr 21, 2026
Conformational entropy in antibodies is inversely proportional to antigen affinity
- antibodies/recognition
- biophysics/dynamics
Apr 21, 2026
Correct CDRH3 prediction is necessary but insufficient for correct Ab-Ag docking
- prediction/complexes
- prediction/antibody-structure
Apr 21, 2026
Correct antibody-antigen prediction in AF3 and related models is partially determined by training set similarity
- evidence/generalization
- prediction/complexes
Apr 21, 2026
Dropout improves AF2 multimer prediction, including for antibody-antigen complexes
- prediction/complexes
- inference/sampling-and-search
Apr 21, 2026
Generic PLMs outperform antibody-specific PLMs on zero-shot predictions of affinity changes
- prediction/binding
- inference/feature-extraction
Apr 21, 2026
Germline precursors bind to their antigens with nanomolar affinity
- antibodies/maturation
- antibodies/recognition
Apr 21, 2026
Increasing diffusion samples is sufficient to yield correctly predicted antibody-antigen complexes
- prediction/complexes
- inference/sampling-and-search
Apr 21, 2026
LM-based antibody structure prediction methods work equally well with generic and Ab PLMs
- prediction/antibody-structure
- inference/feature-extraction
Apr 21, 2026
Lower antibody affinity is desirable for minimizing on-target off-tumor binding
- cell-biology/therapeutic-mechanisms
- antibodies/recognition
Apr 21, 2026
Mutations obtained by antibodies during affinity maturation show epistasis in biophysical properties but not binding
Apr 21, 2026
Nanobodies and Abs target different epitopes but with similar accessibility and number of contacts
- antibodies/nanobodies
- antibodies/recognition
Apr 21, 2026
Nanobody framework residues more likely to be part of paratope than those of antibodies
- antibodies/nanobodies
- antibodies/recognition
Apr 21, 2026
No one-size-fits-all approach to scoring antibody structures
Apr 21, 2026
PAE weakly correlates with Ab-Ag binding
- prediction/confidence
- prediction/binding
Apr 21, 2026
Structure-based methods outperform sequence-based methods on antigen-dependent antibody design
- design/antibodies
- design/inverse-folding
Apr 21, 2026
The structures of nanobodies with kinked CDR3 loops are more difficult to predict in complex with their antigen
- prediction/complexes
- antibodies/nanobodies
Apr 21, 2026
Up to 22 percent of antigen binding residues are outside CDRs
- antibodies/recognition
- antibodies/architecture
Apr 21, 2026
Variable regions can adopt multiple interchain orientations, and these are difficult to predict
Apr 21, 2026
Diffusion-based structure prediction can be steered by modifying the conditioning embeddings rather than the latent space, and such embeddings can be used for subsequent iterations
Apr 21, 2026
Essential proteins are more thermostable and over-expressed than non-essential proteins
- evolution/selection
- biophysics/stability
Apr 21, 2026
Glutamate- and lysine-rich designs can fail due to expression failures resulting from adenosine-rich sequences
- design/developability
- design/inverse-folding
Apr 21, 2026
Inference-time scaling of de novo designed proteins is more effective for harder targets
- inference/sampling-and-search
- design/backbones
Apr 21, 2026
Not all sequences with improved activity have plausible evolutionary paths via stepwise introduction of mutations
- evolution/mutation-effects
- design/directed-evolution
Apr 21, 2026
Using evolutionary information for label smoothing during inverse folding training leads to improved fitness prediction
Apr 21, 2026
A subset of residues is sufficient to predict protein dG
- prediction/stability-expression
Apr 21, 2026
AA identity can be determined from structural context alone
- design/inverse-folding
Apr 21, 2026
AF2 RMSD to ground truth correlated with RMSD between conformations
- prediction/ensembles
- prediction/confidence
Apr 21, 2026
AF3-generation methods can be substantially shrunk without sacrificing prediction quality, except on PPIs
- model-design/architectures
- prediction/complexes
Apr 21, 2026
Ab finetuning improves naturalness of inverse folding designs
Apr 21, 2026
Abs modeled using MD are better for featurization than those modeled using homology modeling
- prediction/stability-expression
- biophysics/molecular-simulation
Apr 21, 2026
Accounting for substitution likelihood improves variant effect prediction relative to log-likelihood
- prediction/variant-effects
- evolution/selection
Apr 21, 2026
Activation-induced deaminase
- antibodies/maturation
- cell-biology/immune-signaling
Apr 21, 2026
Adding noise sometimes leads to greater diversity during inverse folding inference
- design/inverse-folding
- inference/sampling-and-search
Apr 21, 2026
Adding noise while training non-Ab inverse folding models improves self-consistency while worsening sequence recovery
Apr 21, 2026
Adding structural adaptors to language models leads to improvements in thermostability prediction compared to structure-based NNs alone
Apr 21, 2026
Affinity maturation also selects for lower self-association
- antibodies/maturation
- design/developability
Apr 21, 2026
Affinity maturation of de novo minibinders introduces mutations far from the active site
- design/binders
- design/directed-evolution
Apr 21, 2026
All-atom protein structure prediction methods are unable to model PPIs with large buried surface areas
- prediction/complexes
- evidence/generalization
Apr 21, 2026
All-atom scoring functions, but not backbone-only scoring functions, correlate with protein stability
- prediction/stability-expression
Apr 21, 2026
Allelic exclusion and inclusion
- antibodies/repertoires
- antibodies/architecture
Apr 21, 2026
AlphaFlow and ESMFlow
- prediction/ensembles
- model-design/generative-models
Apr 21, 2026
AlphaFold2 and RosettaFold sometimes produce different conformations
- prediction/ensembles
Apr 21, 2026
AlphaFold2 generalizes to unseen secondary structures and topologies
- evidence/generalization
- prediction/structure
Apr 21, 2026
AlphaFold2 outperforms ESMFold at distinguishing designable and undesignable protein backbones
- evidence/design-validation
- design/backbones
Apr 21, 2026
AlphaFold2 predicts Apo conformations with poorer RMSD than holo conformations
- prediction/ensembles
Apr 21, 2026
AlphaFold2 predicts ground states more often
- prediction/ensembles
Apr 21, 2026
AlphaFold2 predicts the holo form of proteins in most cases
- prediction/ensembles
Apr 21, 2026
AlphaFold3 learns intramolecular interactions faster than intermolecular interactions
- training/pretraining-and-scaling
- prediction/complexes
Apr 21, 2026
AlphaFold3 performs comparably to AlphaFold2 when predicting multiple conformations of fold-switching proteins
- prediction/ensembles
Apr 21, 2026
AlphaFold3 universally predicts the active state of ligand-bound GPCRs, even when the ligand is an antagonist
- prediction/ensembles
- evidence/generalization
Apr 21, 2026
AlphaFold3 uses cross-distillation from AlphaFold2 to avoid hallucinating secondary structure in low-pLDDT regions
- evidence/datasets
- prediction/structure
Apr 21, 2026
AlphaMissense
- prediction/variant-effects
- training/fine-tuning
Apr 21, 2026
Alternate conformations can be sampled by inverting AlphaFold2 and backpropagating along specific user-picked collective variables
- prediction/ensembles
- inference/guidance
Apr 21, 2026
Alternate conformations can be sampled with MSA-based structure prediction methods using custom PDB databases and subsampled MSAs
Apr 21, 2026
Alternate sequence clustering schemes outperform uniform sampling when training protein language models
- evidence/datasets
- training/pretraining-and-scaling
Apr 21, 2026
Alternative noise schedules improve training of multimodal PLMs
- training/pretraining-and-scaling
- model-design/multimodal
Apr 21, 2026
Ancestral enzymes are not always more generalist
- evolution/ancestry
- biophysics/catalysis
Apr 21, 2026
Ancestral sequence reconstruction outperforms generative ML methods on functional enzyme design
Apr 21, 2026
Ancestrally reconstructed enzymes show greater expression than extant enzymes
- evolution/ancestry
- design/developability
Apr 21, 2026
Ancestrally reconstructed sequences are more thermostable than extant sequences
- evolution/ancestry
- biophysics/stability
Apr 21, 2026
Anti-nanobody immunogenic reactions disproportionately target their C-termini
- design/developability
- antibodies/nanobodies
Apr 21, 2026
Antibodies and nanobodies can desymmetrize proteins for cryoEM
- design/modular-proteins
- antibodies/nanobodies
Apr 21, 2026
Antibodies can be humanized using AlphaFold2 models
- design/developability
- design/antibodies
Apr 21, 2026
Antibodies from naive B cells are more polyreactive
Apr 21, 2026
Antibody LMs outperform generic PLMs on intrafamily thermostability prediction
- prediction/stability-expression
- inference/feature-extraction
Apr 21, 2026
Antibody PLMs match or outperform generic PLMs on specificity prediction
- prediction/binding
- inference/feature-extraction
Apr 21, 2026
Antibody developability parameters predicted from sequence are more correlated with each other than those predicted from structure
- prediction/stability-expression
- design/developability
Apr 21, 2026
Antibody developability predictions are relatively stable to single point mutations
- prediction/stability-expression
- design/developability
Apr 21, 2026
Antibody developability predictions from computational models are sensitive to the structure prediction method
Apr 21, 2026
Antibody humanization
- design/developability
- design/antibodies
Apr 21, 2026
Antibody language models are able to distinguish correctly and incorrectly paired sequences, but are sensitive to scale and training dataset size
- antibodies/repertoires
- model-analysis/representation-geometry
Apr 21, 2026
Antibody numbering
- antibodies/architecture
Apr 21, 2026
Antibody-drug conjugates
- cell-biology/therapeutic-mechanisms
- design/modular-proteins
Apr 21, 2026
Arginines have a greater propensity to form protein-protein interactions than lysines
- biophysics/interactions
- evolution/selection
Apr 21, 2026
Attention matrices from antibody LMs can be used for paratope prediction
- prediction/binding
- model-analysis/interpretability
Apr 21, 2026
Attention matrices of antibody language models do not correspond to contacts
- model-analysis/interpretability
- prediction/antibody-structure
Apr 21, 2026
Autoimmune diseases involve the constant activation and reactivation of new B cells
- cell-biology/immune-signaling
- antibodies/maturation
Apr 21, 2026
Averaging logits from multiple sources can improve fitness prediction
Apr 21, 2026
B-cell proliferation is driven by both antigen binding and surface expression
- antibodies/maturation
- design/developability
Apr 21, 2026
B-cell receptors
- cell-biology/immune-signaling
Apr 21, 2026
BERT models trained on sequence clusters outperform those trained on all data
- evidence/datasets
- training/pretraining-and-scaling
Apr 21, 2026
BERT-based inverse folding models generate overly repetitive sequences
- design/inverse-folding
- training/objectives-and-optimization
Apr 21, 2026
BRIL
- design/modular-proteins
Apr 21, 2026
Backbone diffusion works best on small protein sizes
- design/backbones
- evidence/design-validation
Apr 21, 2026
Backbone geometry of valine and isoleucine can be detected by inverse folding methods without sequence context
- design/inverse-folding
Apr 21, 2026
Base PLMs must usually be fine-tuned to generate functionally active sequences
- design/sequence-generation
- training/fine-tuning
Apr 21, 2026
Beneficial mutations of enzymes far from active site improve substrate binding and product release
Apr 21, 2026
Biophysical models trained on single and double point mutations can predict effects of three or more mutations on stability
Apr 21, 2026
Biparatopic antibodies can induce clustering beyond what is observed in monoparatopic antibodies
- cell-biology/therapeutic-mechanisms
- biophysics/interactions
Apr 21, 2026
Bispecific antibodies
- cell-biology/therapeutic-mechanisms
- design/modular-proteins
Apr 21, 2026
Broadly neutralizing antibodies
- antibodies/recognition
- antibodies/maturation
Apr 21, 2026
CDR representations segregate into distinct clusters
- model-analysis/representation-geometry
- antibodies/architecture
Apr 21, 2026
CDR3 mediates heavy-light chain pairing preferences by making contacts to framework residues
- antibodies/architecture
Apr 21, 2026
CDRH3 conformation is affected by interdomain orientation
- antibodies/architecture
- biophysics/dynamics
Apr 21, 2026
CDRH3 is shorter in mice and therapeutic Abs
- antibodies/repertoires
- antibodies/architecture
Apr 21, 2026
CUMAb
- design/developability
- design/antibodies
Apr 21, 2026
Categorical Jacobian method
- prediction/structure
- model-analysis/interpretability
Apr 21, 2026
Circulating antibodies can undergo further affinity maturation against new antigen variants
- antibodies/maturation
Apr 21, 2026
Classifiers can be used as predictors of continuous properties
- prediction/binding
Apr 21, 2026
Closely related protein language models have diverging performance on specific domains
- evidence/generalization
Apr 21, 2026
Closely-related enzymes can have disparate catalytic efficiencies that weakly correlate with sequence similarity
- biophysics/catalysis
- prediction/function
Apr 21, 2026
Complementarity-determining regions
- antibodies/architecture
- antibodies/recognition
Apr 21, 2026
Composition of negative data affects rule discovery in binary classification tasks
Apr 21, 2026
Computational models of proteins fit NMR data better than models designed using classical approaches
- prediction/structure
- evidence/measurements
Apr 21, 2026
Conformationally stabilizing mutations can be identified from sequence clusters that are predicted to fold into that conformation when passed to AlphaFold2
Apr 21, 2026
Contrastive learning improves segregation of functional clusters
Apr 21, 2026
Cooperative unfolding is not correlated with overall protein stability across proteins with the same fold
- biophysics/folding
- biophysics/stability
Apr 21, 2026
Correct protomer structure prediction is necessary but insufficient for accurate protein-protein docking
- prediction/complexes
Apr 21, 2026
Coupling sidechain and main chain prediction or design does not always lead to improvements
- design/backbones
- design/sequence-generation
Apr 21, 2026
Cytokines can be added to antibody CDRs
- design/modular-proteins
- design/antibodies
Apr 21, 2026
DL ligand docking methods generate unrealistic poses
- prediction/ligand-docking
- evidence/generalization
Apr 21, 2026
DL structure prediction methods cannot predict CDR conformations unseen during training
- evidence/generalization
- prediction/antibody-structure
Apr 21, 2026
DL structure tools introduce chiral errors such as swapped chiral centers, D-amino acids, and cis-amide bonds
- prediction/structure
- prediction/antibody-structure
Apr 21, 2026
DNA substitutions during somatic hypermutation are spatially clustered
- antibodies/maturation
- evolution/mutation-effects
Apr 21, 2026
De novo designed backbones can increase the size of proteins for cryo-EM structure determination
- design/modular-proteins
- design/backbones
Apr 21, 2026
De novo designed proteins have multiphasic folding pathways
- biophysics/folding
- design/backbones
Apr 21, 2026
De novo designed proteins with alpha helices are easier to predict than those with other secondary structures
Apr 21, 2026
Deep learning methods cannot generalize T-cell receptor binding to new epitopes
- evidence/generalization
- prediction/binding
Apr 21, 2026
Deep learning methods outperform classical methods on sidechain prediction
- prediction/structure
Apr 21, 2026
Designed sequences of flexible proteins are dependent on conformation
- design/inverse-folding
- biophysics/dynamics
Apr 21, 2026
DiffDock
- prediction/ligand-docking
- model-design/generative-models
Apr 21, 2026
Differences in thermostability of structurally similar proteins are due to differences in unfolding rates
- biophysics/folding
- biophysics/stability
Apr 21, 2026
Different CDRL3 lengths are found in functional and dysfunctional antibodies
- antibodies/architecture
- design/developability
Apr 21, 2026
Different antibody lineages can undergo convergent evolution during affinity maturation
- antibodies/maturation
- antibodies/repertoires
Apr 21, 2026
Diffusion outperforms hallucination when using AF3-generation protein structure prediction methods
Apr 21, 2026
Diffusion-based protein design methods undersample structural diversity in specific topologies
- design/backbones
- evidence/datasets
Apr 21, 2026
Diffusion-based protein structure prediction methods double as energy methods comparable to traditional force fields
- prediction/confidence
- model-design/generative-models
Apr 21, 2026
Diffusion-based structure prediction can sometimes model conformational ensembles
- prediction/ensembles
- model-design/generative-models
Apr 21, 2026
Discrete structure tokens are worse than continuous representations for structural alignment tasks
- evolution/homology
- model-design/multimodal
Apr 21, 2026
Display methods
- design/directed-evolution
- evidence/measurements
Apr 21, 2026
Distance between PLM representations of two proteins correlates with functional dissimilarity
- model-analysis/representation-geometry
- prediction/function
Apr 21, 2026
Distance between averaged PLM embeddings does not correlate with structural difference
- model-analysis/representation-geometry
- prediction/structure
Apr 21, 2026
Distillation
- evidence/datasets
- training/pretraining-and-scaling
Apr 21, 2026
Diverse MSAs are better for training structure prediction neural networks than random MSAs
- evidence/datasets
- prediction/structure
Apr 21, 2026
DockQ
- evidence/measurements
- prediction/complexes
Apr 21, 2026
ESM-IF representations can predict if residue binds ligands
- prediction/binding
- inference/feature-extraction
Apr 21, 2026
ESM-IF
- design/inverse-folding
Apr 21, 2026
ESM
- training/pretraining-and-scaling
- inference/feature-extraction
Apr 21, 2026
ESMFold
- prediction/structure
Apr 21, 2026
EVE
- prediction/variant-effects
- evolution/selection
Apr 21, 2026
Effect of missense mutations on folding is greater than on binding
Apr 21, 2026
Effectiveness of the BLOSUM62 for antibody affinity maturation varies by molecule
- design/antibodies
- design/directed-evolution
Apr 21, 2026
Encoder-decoder PLM embeddings are better able to distinguish nearest neighbors than encoder-only PLMs
- model-analysis/representation-geometry
Apr 21, 2026
Endocytosis
- cell-biology/immune-signaling
- cell-biology/therapeutic-mechanisms
Apr 21, 2026
Energy-based models
- model-design/generative-models
- biophysics/thermodynamics
Apr 21, 2026
Enhanced diffusion with metadynamics-like potentials can sometimes convergence slower than unbiased diffusion
- prediction/ensembles
- inference/guidance
Apr 21, 2026
Ensembles can be modeled by structure prediction NNs using experimental data via guidance sampling
Apr 21, 2026
Enzymes can be miniaturized with Monte Carlo sampling and embedding similarity of catalytic residues
- design/enzymes
- inference/sampling-and-search
Apr 21, 2026
Epistasis in antibodies during affinity maturation can be driven by CDR rearrangements
Apr 21, 2026
Epistasis in enzymes can arise from changes in the rate-limiting step of the catalytic cycle
- evolution/mutation-effects
- biophysics/catalysis
Apr 21, 2026
Epistasis is rare during evolution
- evolution/mutation-effects
- evolution/selection
Apr 21, 2026
Epistasis more likely to occur in buried residues than surface residues
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
Equilibrium dynamics of fold switching proteins change during evolution
- evolution/ancestry
- biophysics/dynamics
Apr 21, 2026
Evo-velocity
- evolution/selection
- model-analysis/interpretability
Apr 21, 2026
Evoformer
- model-design/architectures
- prediction/structure
Apr 21, 2026
Evolutionary pathway between two folds identified using AlphaFold2 and ancestral sequence reconstruction
- evolution/ancestry
- biophysics/dynamics
Apr 21, 2026
FAPE can be extended to ligands
- training/objectives-and-optimization
- prediction/ligand-docking
Apr 21, 2026
FcRn
- cell-biology/therapeutic-mechanisms
- cell-biology/immune-signaling
Apr 21, 2026
Features for antibody property prediction derived from MD simulations outperform those from language models and static structures
Apr 21, 2026
Fine-tuning AlphaFold for MHC prediction by adding Evoformer layers improves RMSD and pLDDT
- training/fine-tuning
- prediction/structure
Apr 21, 2026
Fine-tuning base models on test cases can improve the performance of variant effect and structure prediction
Apr 21, 2026
Fitness prediction
- prediction/variant-effects
- evolution/mutation-effects
Apr 21, 2026
Flow matching
- model-design/generative-models
Apr 21, 2026
Frame aligned point error
- training/objectives-and-optimization
- prediction/structure
Apr 21, 2026
Framework region
- antibodies/architecture
Apr 21, 2026
Free energy perturbation
Apr 21, 2026
Function prediction
- prediction/function
Apr 21, 2026
GRAB motifs
- antibodies/recognition
- antibodies/repertoires
Apr 21, 2026
Geometric Vector Perceptrons
- model-design/architectures
- design/inverse-folding
Apr 21, 2026
Germline
- antibodies/repertoires
Apr 21, 2026
Glycosylation
- design/developability
Apr 21, 2026
Guided sequence-only diffusion outperforms autoregressive fine-tuned PLMs on designing high-fitness sequences
Apr 21, 2026
Half of all disease-causing missense variants due to disruption of protein stability
Apr 21, 2026
Hallucination outperforms diffusion on protein refolding accuracy, particularly among larger proteins
- design/backbones
- evidence/design-validation
Apr 21, 2026
Heavy and light chains of de novo designed antibodies with the same framework and binding mode can be exchanged
- design/antibodies
- antibodies/architecture
Apr 21, 2026
Heavy chains
- antibodies/architecture
Apr 21, 2026
High-confidence predictions from protein language models co-cluster together in embedding space and correlate with performance on variant effect prediction tasks
Apr 21, 2026
High-stability substructures fold first
- biophysics/folding
Apr 21, 2026
Highly thermostable sequences make better starting points to evolve new functions
Apr 21, 2026
Homooligomeric proteins can undergo many stoichiometric changes in the course of evolution
- evolution/selection
- biophysics/interactions
Apr 21, 2026
Humanization can destabilize antibodies
Apr 21, 2026
Hybrid sequence-structure models
- model-design/multimodal
Apr 21, 2026
IPA-based structure prediction neural networks train sequentially along three spatial axes
- training/pretraining-and-scaling
- prediction/structure
Apr 21, 2026
IgFold
- prediction/antibody-structure
- model-design/architectures
Apr 21, 2026
ImmuneBuilder
- prediction/antibody-structure
Apr 21, 2026
Improvements in antigen binding are correlated with decreases in expression
- design/antibodies
- design/developability
Apr 21, 2026
Including confidence improves DL-based docking
- prediction/ligand-docking
- training/objectives-and-optimization
Apr 21, 2026
Including ligands restrains design of catalytically active residues
- design/inverse-folding
- design/enzymes
Apr 21, 2026
Including sequences from multiplexed ancestral sequence reconstruction improves PLM training
Apr 21, 2026
Including structure prediction confidence while training inverse folding improves sequence diversity but not sequence recovery
Apr 21, 2026
Inclusion of adjacent loops can improve CDRH3 structural modeling
Apr 21, 2026
Increasing RosettaFold embedding dimension improves structure prediction
- model-design/architectures
- prediction/structure
Apr 21, 2026
Indels allows proteins to access phenotypes inaccessible to missense mutations
- evolution/mutation-effects
- evolution/selection
Apr 21, 2026
Individual D genes are evolutionarily selected for well-defined antigen-binding capabilities
- antibodies/repertoires
- antibodies/recognition
Apr 21, 2026
Insufficient ddG data on Abs for training
- prediction/binding
- evidence/datasets
Apr 21, 2026
Interferons
- cell-biology/immune-signaling
Apr 21, 2026
Interleukins
- cell-biology/immune-signaling
Apr 21, 2026
Invariant point attention
- model-design/architectures
Apr 21, 2026
Inverse folding NNs are better predictors of equilibrium dynamics than protein folding NNs
- prediction/ensembles
- design/inverse-folding
Apr 21, 2026
Inverse folding can be improved by incorporating larger PLMs
- design/inverse-folding
- model-design/multimodal
Apr 21, 2026
Inverse folding can generate more stable sequences when jointly run alongside a protein folding model
- design/inverse-folding
- biophysics/stability
Apr 21, 2026
Inverse folding from MD simulations
Apr 21, 2026
Inverse folding generalizes to unseen structures better than PLMs
- evidence/generalization
- design/inverse-folding
Apr 21, 2026
Inverse folding models can distinguish between and design backbones belonging to transmembrane and soluble beta-barrels
- design/inverse-folding
Apr 21, 2026
Inverse folding models trained on all proteins generate sequences that are equally thermostable as those generated by models trained only on thermophilic proteins
Apr 21, 2026
Inverse folding of CDRs benefits from span masking during training
Apr 21, 2026
Inverse folding sequence perplexities correlate with Rosetta energies, forward folding TM-scores, and sequence recovery
Apr 21, 2026
Inverse folding sequence recovery can be improved by simultaneously learning side chain packing
- design/inverse-folding
- training/objectives-and-optimization
Apr 21, 2026
Iterative structure prediction outperforms all-at-once structure prediction
- prediction/structure
- inference/sampling-and-search
Apr 21, 2026
Kappa light chains are far more common in antibody therapeutics
- antibodies/architecture
- design/developability
Apr 21, 2026
Kappa subtype restricts CDRH3 dynamics
- antibodies/architecture
- biophysics/dynamics
Apr 21, 2026
Kinases
- biophysics/catalysis
Apr 21, 2026
Kinks in nanobody CDR3 loops are correctly predicted by structure prediction methods
Apr 21, 2026
Lambda light chains are less thermostable
- biophysics/stability
- antibodies/architecture
Apr 21, 2026
Lambda light chains are more hydrophobic
- design/developability
- antibodies/architecture
Apr 21, 2026
Language models can be infused with structure via low-rank adapter layers
Apr 21, 2026
Language-based protein folding NNs predict novelty with far lower confidence than MSA-based protein folding NNs
- evidence/generalization
- prediction/structure
Apr 21, 2026
Larger PLMs are better at homolog detection
- evolution/homology
- training/pretraining-and-scaling
Apr 21, 2026
Larger PLMs are better at thermostability prediction
- prediction/stability-expression
- training/pretraining-and-scaling
Apr 21, 2026
Larger PLMs generate more novel sequences from more sparsely populated protein families
- design/sequence-generation
- training/pretraining-and-scaling
Apr 21, 2026
Light chain CDR charges can shorten half-life by slowing release from FcRn
Apr 21, 2026
Light chain coherence
- antibodies/repertoires
- antibodies/architecture
Apr 21, 2026
Light chain determines autospecificity
Apr 21, 2026
Light chain representations affected by heavy chain, but not vice versa
- model-analysis/interpretability
- antibodies/architecture
Apr 21, 2026
Logistic regression outperforms fine-tuned LMs on finding point mutations from NGS data
- design/directed-evolution
- evidence/design-validation
Apr 21, 2026
Loops from one V-region chain affect the conformations and dynamics of loops in the other chain
- antibodies/architecture
- biophysics/dynamics
Apr 21, 2026
Loss functions
- training/objectives-and-optimization
Apr 21, 2026
Low-pLDDT hallucinated models from AF3-generation protein structure predictors are designable
Apr 21, 2026
MD trajectories reproduce experimental dG values to within 1 kcal per mol
- biophysics/molecular-simulation
- prediction/stability-expression
Apr 21, 2026
ML improves directed evolution when navigating fitness landscapes with greater non-magnitude epistasis
- design/directed-evolution
- evolution/mutation-effects
Apr 21, 2026
ML models must trade off bias and variance
- prediction/variant-effects
- prediction/confidence
Apr 21, 2026
ML models trained exclusively on experimental structures are less effective on computational models
Apr 21, 2026
ML optimized Ab libraries have more mutations than PSSM libraries
- design/antibodies
- design/directed-evolution
Apr 21, 2026
MPNN-like design profiles may be the source of high thermostability
- design/inverse-folding
- biophysics/stability
Apr 21, 2026
MSA-based PLMs outperform generic PLMs on predicting structure and stability
Apr 21, 2026
MSA-based structure prediction models can be retrained as pathogenicity prediction models by upweighting BERT losses
Apr 21, 2026
MSA-based structure predictions outperform PLM-based methods
- prediction/structure
Apr 21, 2026
Major histocompatibility complex II
- cell-biology/immune-signaling
Apr 21, 2026
Masked PLMs are more sensitive to training imbalances than autoregressive PLMs
- evidence/datasets
- training/pretraining-and-scaling
Apr 21, 2026
Mean-pooled embeddings outperform other zero-shot approaches for transfer learning of PLMs using the full sequence
- inference/feature-extraction
Apr 21, 2026
Membrane proteins are predicted by PLMs via solvent-exposed hydrophobic residues
- model-analysis/interpretability
- prediction/structure
Apr 21, 2026
Membrane proteins can be bulked up for cryo-EM by inserting designed domains, including those with predefined epitopes for antibodies
- design/modular-proteins
Apr 21, 2026
Metal-binding enzymes are more evolutionarily constrained than non-metal-binding enzymes
- evolution/selection
- biophysics/catalysis
Apr 21, 2026
Missense mutation effects are predictive of effects of indels
- evolution/mutation-effects
- prediction/variant-effects
Apr 21, 2026
Molecular glues
- cell-biology/therapeutic-mechanisms
- biophysics/interactions
Apr 21, 2026
Monomeric AF2 is better able to distinguish positive and negative PPIs than AF2-multimer
Apr 21, 2026
Most ML quality metrics cannot effectively predict enzyme activity after controlling for similarity to native
Apr 21, 2026
Most camelid-derived nanobodies have high sequence identity with human IGHV3
- antibodies/nanobodies
- design/developability
Apr 21, 2026
Most commercially available tool antibodies are nonspecific
- antibodies/recognition
- evidence/measurements
Apr 21, 2026
Most homo-oligomeric complexes are symmetric
- biophysics/interactions
Apr 21, 2026
Most mutations with no effect on stability are found near the surface of proteins
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
Most point mutations that antibodies obtain during affinity maturation are the result of single nucleotide changes
- antibodies/maturation
- evolution/mutation-effects
Apr 21, 2026
Most possible alpha-beta topologies unexplored by nature
- evolution/selection
- design/backbones
Apr 21, 2026
Most random mutants are nonfunctional
- evolution/mutation-effects
- design/directed-evolution
Apr 21, 2026
Most viral proteins are structurally distinct from eukaryotic proteins
- evolution/selection
- evolution/homology
Apr 21, 2026
Multi-sequence protein language models
- model-design/architectures
- inference/conditioning
Apr 21, 2026
Multimodal sequence-structure guidance outperforms unimodal guidance in cases where both need to be designed
Apr 21, 2026
Multiple sequence alignments
- evolution/homology
Apr 21, 2026
Multistate Bennett acceptance ratio can be used to reweigh one or more samples from a guided diffusion trajectory
- biophysics/thermodynamics
- inference/guidance
Apr 21, 2026
Mutation memory half life
- evolution/mutation-effects
Apr 21, 2026
N- and C-termini that are close in space are correlated with correct protein folding
- biophysics/folding
Apr 21, 2026
Nanobodies designed de novo from sequence are more flexible than either those designed from structure or those occurring in nature
Apr 21, 2026
Nanobodies
- antibodies/nanobodies
- antibodies/architecture
Apr 21, 2026
Nanobody CDR1 and CDR2 do not conform to the canonical clustering patterns found in traditional antibodies
- antibodies/nanobodies
- antibodies/architecture
Apr 21, 2026
Nanobody CDRH3 loops are longer but more compact
- antibodies/nanobodies
- antibodies/architecture
Apr 21, 2026
Nanobody FR2 is more hydrophilic than that of antibodies
- antibodies/nanobodies
- design/developability
Apr 21, 2026
Natural sequences have higher pTM but lower pLDDT than de novo sequences
- evidence/design-validation
- prediction/confidence
Apr 21, 2026
Nearly half of CATH domains do not pass designability filters used to evaluate protein backbone design performance
- evidence/design-validation
- design/backbones
Apr 21, 2026
Net charge of CDRs strongly predicts nonspecific binding
- design/developability
- antibodies/recognition
Apr 21, 2026
Neutrality correlates with fold-level thermostability
- biophysics/stability
- evolution/selection
Apr 21, 2026
Next-generation sequencing
- evidence/measurements
- antibodies/repertoires
Apr 21, 2026
No one-size-fits-all best approach to motif scaffolding protein design
- design/backbones
- evidence/design-validation
Apr 21, 2026
Normalizing flows
- model-design/generative-models
Apr 21, 2026
Number of sequences with 90% homology is a greater predictor of viral variant effect prediction than alignment depth
Apr 21, 2026
Observable Antibody Space
- evidence/datasets
- antibodies/repertoires
Apr 21, 2026
OmegaFold
- prediction/structure
- prediction/antibody-structure
Apr 21, 2026
Optimal scaling of antibody LMs improves prediction of masked CDRH3 residues but not framework residues
- training/pretraining-and-scaling
- antibodies/architecture
Apr 21, 2026
Optimal transport outperforms mean-pooling on property prediction tasks
- inference/feature-extraction
Apr 21, 2026
Ordered SAEs are more steerable than top-k SAEs
- model-analysis/interpretability
- inference/guidance
Apr 21, 2026
PD-1
- cell-biology/immune-signaling
- cell-biology/therapeutic-mechanisms
Apr 21, 2026
PD-L1
- cell-biology/immune-signaling
- cell-biology/therapeutic-mechanisms
Apr 21, 2026
PLM attention maps from specific heads can be used to predict allosteric networks
- model-analysis/interpretability
- biophysics/dynamics
Apr 21, 2026
PLM attention matrices correspond to 3D contacts
- model-analysis/interpretability
- prediction/structure
Apr 21, 2026
PLM embeddings contain enough information to align proteins without fine-tuning
- evolution/homology
- inference/feature-extraction
Apr 21, 2026
PLM embeddings fine-tuned using contrastive learning outperform other representations in drug-target interaction prediction
Apr 21, 2026
PLM-based sequence searches outperform sequence- and matches structure-based search methods
- evolution/homology
- inference/feature-extraction
Apr 21, 2026
PLM-derived antibody representations can distinguish engineered from human-derived Abs
- model-analysis/representation-geometry
- design/antibodies
Apr 21, 2026
PLM-designed sequences match the distribution of fitness values, lengths, and structure prediction confidence of natural sequences
- design/sequence-generation
- evidence/design-validation
Apr 21, 2026
PLMs are better at reproducing sequence-based developability predictions than structure-based predictions
- prediction/stability-expression
- inference/feature-extraction
Apr 21, 2026
PLMs are biased toward germline antibodies
- evidence/datasets
- antibodies/repertoires
Apr 21, 2026
PLMs cluster antibodies from the same repertoire by V gene and maturation status
Apr 21, 2026
PLMs downweigh probability of sequences with multiple mutations
- prediction/variant-effects
- evidence/generalization
Apr 21, 2026
PLMs generate more divergent functional proteins than Potts models
- design/sequence-generation
- evidence/design-validation
Apr 21, 2026
PROTACs
- cell-biology/therapeutic-mechanisms
Apr 21, 2026
Paired antibody LMs outperform equivalent unpaired LMs on sequence recovery for various regions
Apr 21, 2026
Partial diffusion of de novo designed binders and VHHs improves refoldability and designability scores
Apr 21, 2026
Partial diffusion of protein structures reduces structural complexity
- design/backbones
Apr 21, 2026
Partial structure diffusion can make de novo backbones more designable
- design/backbones
- evidence/design-validation
Apr 21, 2026
Precision decreases and recall increases as variant effect prediction focuses on top-performing point mutations
- prediction/variant-effects
- evidence/measurements
Apr 21, 2026
ProGen
- training/pretraining-and-scaling
- design/sequence-generation
Apr 21, 2026
Proline content correlates with high unfolding cooperativity but not overall stability
- biophysics/folding
- biophysics/stability
Apr 21, 2026
Protein abundance is inversely correlated with its propensity to form nonfunctional interactions
- evolution/selection
- biophysics/interactions
Apr 21, 2026
Protein backbone diffusion models undersample loop-rich and alpha-beta domains and functional motifs
- design/backbones
- evidence/datasets
Apr 21, 2026
Protein backbones designed by diffusion, but not by language models, have more secondary structure
- design/backbones
- design/sequence-generation
Apr 21, 2026
Protein backbones designed using diffusion, but not sequence-based models, have fewer beta sheets
- design/backbones
- design/sequence-generation
Apr 21, 2026
Protein dynamics
- biophysics/dynamics
- biophysics/ensemble-analysis
Apr 21, 2026
Protein ensemble prediction methods do not generate conformations that are amenable to PPI docking
- prediction/ensembles
- prediction/complexes
Apr 21, 2026
Protein folding duration scales with the inverse square root of its length
- biophysics/folding
Apr 21, 2026
Protein folding neural networks can be combined with VAEs for conformational modeling
- prediction/ensembles
- model-design/generative-models
Apr 21, 2026
Protein folding neural networks can be subject to adversarial attacks
- evidence/generalization
- prediction/structure
Apr 21, 2026
Protein folding neural networks do not extrapolate to new ligand binding sites
- evidence/generalization
- prediction/structure
Apr 21, 2026
Protein folding neural networks do not learn the physics of protein folding for most proteins
Apr 21, 2026
Protein folding neural networks make local optimizations in the absence of coevolutionary information
- model-analysis/interpretability
- prediction/structure
Apr 21, 2026
Protein language model embeddings are more predictive of homology than catalytic efficiency
Apr 21, 2026
Protein language models and PLM-based structure prediction generalize to de novo designed proteins
Apr 21, 2026
Protein language models are able to predict epistasis in a zero-shot setting following a nonlinear transform
- prediction/variant-effects
- evolution/mutation-effects
Apr 21, 2026
Protein language models are better zero-shot predictors for ranking closely related sequences than distantly related sequences
- prediction/variant-effects
- evidence/generalization
Apr 21, 2026
Protein language models can be steered to design proteins with specific properties
- inference/guidance
- design/sequence-generation
Apr 21, 2026
Protein language models can predict zero-shot which proteins belong to the same species
Apr 21, 2026
Protein language models have a lower bound of training loss regardless of size
- training/pretraining-and-scaling
Apr 21, 2026
Protein language models make equally effective predictions when trained on individual proteins or protein families
- training/pretraining-and-scaling
- prediction/variant-effects
Apr 21, 2026
Protein models designed using inverse folding can be used to supplement training DBs for PLMs and structure prediction models
- evidence/datasets
- training/pretraining-and-scaling
Apr 21, 2026
Protein stability, expression, and resistance to proteolysis are correlated
- biophysics/stability
- design/developability
Apr 21, 2026
Protein structure more conserved than sequence
- evolution/homology
- evolution/selection
Apr 21, 2026
Protein structure prediction and design metrics don't correlate with expression probability
Apr 21, 2026
Protein structure prediction methods are unable to predict the energetics of a conformational landscape unless explicitly trained for that purpose
- prediction/ensembles
- evidence/generalization
Apr 21, 2026
Protein structure prediction methods predict idealized secondary structures in de novo proteins
- prediction/structure
- evidence/design-validation
Apr 21, 2026
Protein thermophilicity has evolved multiple times
- evolution/selection
- biophysics/stability
Apr 21, 2026
Protein-ligand co-folding methods do not generalize beyond their training set
- evidence/generalization
- prediction/ligand-docking
Apr 21, 2026
Protein-ligand co-folding
- prediction/ligand-docking
- prediction/complexes
Apr 21, 2026
Protein-protein interaction interfaces are highly degenerate
- biophysics/interactions
Apr 21, 2026
Protein-protein interaction interfaces promote cotranslational assembly
- biophysics/interactions
- biophysics/folding
Apr 21, 2026
Protein-protein interfaces have looser packing than intra-chain interactions
- biophysics/interactions
Apr 21, 2026
ProteinMPNN and AlphaFold structure update module can be combined into a VAE
- model-design/multimodal
- model-design/generative-models
Apr 21, 2026
ProteinMPNN-derived inverse folding methods underdesign aromatic residues
- design/inverse-folding
- design/binders
Apr 21, 2026
ProteinMPNN
- design/inverse-folding
Apr 21, 2026
ProteinNPT
- prediction/variant-effects
- model-design/architectures
Apr 21, 2026
Proteins designed by diffusion are more compact than those designed by hallucination
- design/backbones
Apr 21, 2026
Proteins with greater folding stability have more intermediate partially unfolded states
- biophysics/folding
- biophysics/stability
Apr 21, 2026
PyMOL
- prediction/structure
Apr 21, 2026
PyRosetta convert mAb structure numbering
- antibodies/architecture
- prediction/antibody-structure
Apr 21, 2026
QM-MM and unbiased MD are insufficient to correctly determine CDRH3 conformation
- prediction/antibody-structure
- biophysics/molecular-simulation
Apr 21, 2026
RMSD is a poor training objective for structure prediction
- training/objectives-and-optimization
- prediction/structure
Apr 21, 2026
Randomly masking residues prior to running ESMFold allows false positives to be identified more effectively
Apr 21, 2026
Recessive genetic diseases more likely to be caused by protein destabilization than dominant genetic diseases
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
Reinforcement learning outperforms fine-tuning in property-guided inverse folding
Apr 21, 2026
Residue conservation and solvent exposure data perform comparably to PLMs at some property prediction tasks
Apr 21, 2026
Ribosomes penalize the unfolded states of proteins
- biophysics/folding
Apr 21, 2026
Running inverse folding in a random order leads to greater sequence recovery than running in a fixed-order
- design/inverse-folding
- inference/sampling-and-search
Apr 21, 2026
Scrambled MSAs outperform single-sequence-prediction with AlphaFold
- prediction/structure
- inference/conditioning
Apr 21, 2026
Search algorithms are more important than language model choice for model-guided protein engineering and design
- inference/sampling-and-search
- design/directed-evolution
Apr 21, 2026
Self-consistency perplexity is correlated with pLDDT
- evidence/design-validation
- prediction/confidence
Apr 21, 2026
Self-distillation exposes more data to Evoformer
- evidence/datasets
- prediction/structure
Apr 21, 2026
Sequence clustering of training data affects variant effect prediction performance by PLMs
- evidence/datasets
- prediction/variant-effects
Apr 21, 2026
Sequence homology composition can affect performance of fine-tuned protein language models for variant effect prediction
Apr 21, 2026
Sequence perplexity and TM-score are negatively correlated when predicting structure using protein language models
- prediction/confidence
- prediction/structure
Apr 21, 2026
Sequence perplexity
- prediction/confidence
- design/sequence-generation
Apr 21, 2026
Sequence recovery in inverse folding models is not correlated with self-consistency of generated designs
- evidence/design-validation
- design/inverse-folding
Apr 21, 2026
Sequence recovery of inverse folding lower for smaller proteins
- design/inverse-folding
Apr 21, 2026
Sequence recovery using inverse folding is lower on NMR structures than crystal structures
- design/inverse-folding
- evidence/measurements
Apr 21, 2026
Sequence-based DL models learn intra-family phylogenetics of protein evolution
- model-analysis/interpretability
- evolution/selection
Apr 21, 2026
Sequence-based variant effect prediction methods generalize to de novo proteins
- evidence/generalization
- prediction/variant-effects
Apr 21, 2026
Sequences with lower log-likelihoods are worse for zero-shot variant effect prediction using PLMs
Apr 21, 2026
Sheets are less designable than helices
- design/inverse-folding
- design/backbones
Apr 21, 2026
Short insertions are common in evolution
- evolution/mutation-effects
- evolution/selection
Apr 21, 2026
SimpleFold
Apr 21, 2026
Simultaneous design of scFv loops predicted to lead to more fit sequences
- design/antibodies
Apr 21, 2026
Single chain variable fragments
- antibodies/architecture
- design/modular-proteins
Apr 21, 2026
Single deletion mutations are better tolerated than larger (2-3 residue) deletions
- evolution/mutation-effects
Apr 21, 2026
Smooth protein folding landscapes are the result of natural selection
- evolution/selection
- biophysics/folding
Apr 21, 2026
Sodium- and proton-coupled transporters share a cation-independent ancestor
- evolution/ancestry
- biophysics/dynamics
Apr 21, 2026
Somatic hypermutation
- antibodies/maturation
- antibodies/repertoires
Apr 21, 2026
Some light chain V-genes are more likely to be found in dysfunctional antibodies
- antibodies/repertoires
- design/developability
Apr 21, 2026
Sources of sequence diversity in antibodies
- antibodies/repertoires
Apr 21, 2026
Sparse autoencoder-derived features do not outperform PLM-derived embeddings for downstream prediction
- inference/feature-extraction
- model-analysis/interpretability
Apr 21, 2026
Sparse autoencoder
- model-analysis/interpretability
- model-design/architectures
Apr 21, 2026
Stability-activity trade-off during enzyme design and evolution is highly local and not global
Apr 21, 2026
Structural classifications can be learned from PLM-based alignment prediction
- evolution/homology
- inference/feature-extraction
Apr 21, 2026
Structural modeling introduces bias toward larger hydrophobic patches in antibodies
Apr 21, 2026
Structure ensemble prediction methods are unable to accurately model subtle conformational differences captured by X-ray crystallography
- prediction/ensembles
- evidence/measurements
Apr 21, 2026
Structure prediction from conformational samplers tends to be worse than those from vanilla structure prediction methods
- prediction/ensembles
Apr 21, 2026
Structure prediction methods undersample the conformational space they find to be high-confidence
- prediction/ensembles
- prediction/confidence
Apr 21, 2026
Structure prediction metrics can be used to predict oligomer stoichiometry
- prediction/complexes
- prediction/confidence
Apr 21, 2026
Structure prediction neural networks predict different conformations of a given sequence depending on if a single chain is predicted versus multiple chains
- prediction/ensembles
- prediction/complexes
Apr 21, 2026
Structure prediction neural networks trained on the same structural dataset are likely to predict the same conformation
- prediction/ensembles
- evidence/datasets
Apr 21, 2026
Structure prediction with AlphaFold improved when starting from an initial guess, rather than default initialization
Apr 21, 2026
Structure-based methods outperform sequence-based methods at zero-shot prediction of binding, whereas the reverse is true for zero-shot prediction of enzymatic activity
- prediction/binding
- prediction/function
Apr 21, 2026
Structure-based methods outperform sequence-based methods on protein stability prediction of point mutants, but not full sequences
- prediction/stability-expression
- inference/feature-extraction
Apr 21, 2026
Structure-derived embeddings outperform sequence-derived embeddings on LM-based alignment
- evolution/homology
- model-design/multimodal
Apr 21, 2026
Subnetworks within protein language models encode specific protein families
- model-analysis/interpretability
- training/pretraining-and-scaling
Apr 21, 2026
Subquadratic scaling for protein backbone diffusion
- model-design/architectures
- design/backbones
Apr 21, 2026
Substitution matrix for inverse folding closely matches BLOSUM62 matrix except proline
- design/inverse-folding
- evolution/selection
Apr 21, 2026
Taking the minimum logit can outperform averaging logits when ensembling models for pathogenicity prediction
Apr 21, 2026
Template CDR retrieval using embeddings
- prediction/antibody-structure
- inference/feature-extraction
Apr 21, 2026
The AlphaFold and ESMFold databases can be mined for alternate conformations
- prediction/ensembles
- evolution/homology
Apr 21, 2026
The PDB already contains implicit information about conformational dynamics based on subtle differences in closely related structures
- evidence/measurements
- biophysics/ensemble-analysis
Apr 21, 2026
The activity profiles of ancestrally reconstructed enzymes are more difficult to predict than extant or ML-generated enzymes
Apr 21, 2026
The confidence metrics of AlphaFold2 are better calibrated than those of AlphaFold3
- prediction/confidence
Apr 21, 2026
The latent space of ESMFold is dicontinuous, but can be smoothened by channel compression
- model-analysis/representation-geometry
- model-design/multimodal
Apr 21, 2026
The majority of missense mutations are destabilizing
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
The solubilization tetrad allows camelid nanobodies to remain soluble in the absence of a light chain
- antibodies/nanobodies
- design/developability
Apr 21, 2026
The structures of shorter de novo designed sequences are easier to predict than those of longer proteins
- prediction/structure
- evidence/design-validation
Apr 21, 2026
There are 4.6 million possible combinations of human antibodies given VDJ and VJ scrambling alone
- antibodies/repertoires
Apr 21, 2026
Thermophilic proteins have more aliphatic and charged residues on surface than mesophilic proteins
- biophysics/stability
- evolution/selection
Apr 21, 2026
Thermostable proteins have more small aliphatic amino acids that facilitate more flexible contacts
- biophysics/stability
- biophysics/dynamics
Apr 21, 2026
Training backbone diffusion models on synthetic data improves designability
- evidence/datasets
- design/backbones
Apr 21, 2026
Training on newly updated databases doesn't guarantee better performance
- evidence/datasets
- training/pretraining-and-scaling
Apr 21, 2026
Training protein structure prediction neural networks on both positive and negative protein-protein interactions improves PPI discrimination
- prediction/binding
- evidence/datasets
Apr 21, 2026
Tranception
- prediction/variant-effects
- training/pretraining-and-scaling
Apr 21, 2026
Transformer
- model-design/architectures
Apr 21, 2026
Transgenic mice with only a single VH gene are nonetheless able to make effective antibodies against many targets
- antibodies/repertoires
- antibodies/recognition
Apr 21, 2026
Transporters can evolve specificity to new substrates with just a single mutation
- design/directed-evolution
- evolution/mutation-effects
Apr 21, 2026
Triangular update can be repurposed for inverse folding
- model-design/architectures
- design/inverse-folding
Apr 21, 2026
Triangular updates are dispensable for structure prediction
- model-design/architectures
- prediction/structure
Apr 21, 2026
Unbalanced composition of sequence data prevents protein fitness from being identifiable from sequence data alone
Apr 21, 2026
Unfolded and nonfunctional isoforms of folded proteins are incorrectly predicted as folded by protein folding neural networks
- evidence/generalization
- prediction/structure
Apr 21, 2026
Unfolding cooperativity is more difficult to predict than protein stability
- prediction/stability-expression
- biophysics/folding
Apr 21, 2026
Unfreezing PLM during structure training improves prediction quality
- training/fine-tuning
- prediction/structure
Apr 21, 2026
Variant effect prediction quality correlates with DMS prediction quality
- prediction/variant-effects
- evidence/measurements
Apr 21, 2026
Variant effect prediction with homology-aware PLMs improves with ensembling of multiple prompts
Apr 21, 2026
Viral receptor-binding domains can bind the same receptor at different locations due to convergent evolution
- evolution/selection
- biophysics/interactions
Apr 21, 2026
ddG predictions calculated from AlphaFold structures are equally accurate as those derived from experimental structures
- prediction/stability-expression
- evidence/generalization
Apr 21, 2026
pLDDT and PAE in protein folding neural networks are correlated
- prediction/confidence
Apr 21, 2026
pLDDT and PAE inversely correlated with protein dynamics in dynamic naturally occurring proteins, but not de novo proteins
- prediction/confidence
- biophysics/dynamics
Apr 21, 2026
pLDDT correlates with number of homologous sequences provided during runtime
- prediction/confidence
- inference/conditioning
Apr 21, 2026
pLDDT is inversely correlated with CDRH3 length
- prediction/confidence
- prediction/antibody-structure
Apr 21, 2026
pTM is less lenient than pLDDT toward structural novelty when predicting alternate conformations
- prediction/confidence
- prediction/ensembles
Apr 21, 2026
Amino acid substitution patterns are dependent on solvent exposure
- evolution/mutation-effects
- evolution/selection
Apr 21, 2026
Amino acid substitution rates are rotamer-dependent
- evolution/mutation-effects
- evolution/selection
Apr 21, 2026
Arginine residues in CDRH3 increases promiscuity
Apr 21, 2026
Flow matching and diffusion perform comparably for biomolecular structure prediction
- prediction/structure
- model-design/generative-models
Apr 21, 2026
Half of pathogenic mutations in intrinsically disordered regions improve binding to partners, or create new binding partners
- evolution/mutation-effects
- biophysics/interactions
Apr 21, 2026
Hybrid protein sequence-structure ML models rely more on sequence when many homologs were available during training
- model-design/multimodal
- evidence/generalization
Apr 21, 2026
Invariant point attention is dispensable to structure prediction
- model-design/architectures
- prediction/structure
Apr 21, 2026
CRISPR-Cas9
- design/directed-evolution
Apr 21, 2026
Cytokines
- cell-biology/immune-signaling
- design/modular-proteins
Apr 21, 2026
Fitness landscapes are locally smooth but globally rugged
- evolution/mutation-effects
Apr 21, 2026
Text-to-structure protein design models outperform text-to-sequence models
- design/backbones
- model-design/multimodal
Apr 21, 2026
Bias-exchange metadynamics
- biophysics/molecular-simulation
Apr 21, 2026
Boltzmann distribution
- biophysics/thermodynamics
Apr 21, 2026
Helmholtz free energy
- biophysics/thermodynamics
Apr 21, 2026
Monte Carlo Tree Search
- inference/sampling-and-search
Apr 21, 2026
NVE ensemble
- biophysics/thermodynamics
- biophysics/molecular-simulation
Apr 21, 2026
NVT ensemble
- biophysics/thermodynamics
- biophysics/molecular-simulation
Apr 21, 2026
Replica-exchange molecular dynamics
- biophysics/molecular-simulation
Apr 21, 2026
A greater proportion of viral sequences are closely related to one another than prokaryotic and eukaryotic sequences, and these sequences get removed during clustering
- evidence/datasets
- evolution/selection
Apr 21, 2026
About 100k datapoints required to train accurate ddG predictor
- prediction/stability-expression
- evidence/datasets
Apr 21, 2026
Accurate fitness landscapes are unnecessary for productively engineering enzymes or proteins
- design/directed-evolution
- evidence/design-validation
Apr 21, 2026
All-atom structure prediction of RNA is driven by memorization
- evidence/generalization
- prediction/complexes
Apr 21, 2026
Alpha carbon-only representations are sufficient for ML force fields
- biophysics/molecular-simulation
- model-design/architectures
Apr 21, 2026
Alpha-helices are more tolerant of mutations than beta-sheets
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
Amino acid propensities in the proteome correlate with differences in metabolic flux
- evolution/selection
Apr 21, 2026
Antibody sequence similarity is not correlated with developability similarity
- design/developability
- prediction/stability-expression
Apr 21, 2026
Axe-like CDRH3 loops
- antibodies/architecture
Apr 21, 2026
Biomolecular diffusion models cannot reproduce the equilibrium dynamics of the simulations they are trained on
Apr 21, 2026
Biophysical models outperform linear models for stability prediction
- prediction/stability-expression
- prediction/variant-effects
Apr 21, 2026
Broadly neutralizing antibodies against HIV evolve in a subset of patients
- antibodies/maturation
- antibodies/recognition
Apr 21, 2026
Broadly neutralizing antibodies are more evolutionarily distant from germline sequences than less polyspecific antibodies
- antibodies/maturation
- evolution/mutation-effects
Apr 21, 2026
Buried nonpolar surface area is a major determinant of whether de novo designed proteins are stable
- biophysics/stability
- design/backbones
Apr 21, 2026
CDRH3 flanking residues more conserved and easier to predict than those in the center
- antibodies/architecture
- prediction/antibody-structure
Apr 21, 2026
CabBCII-10 is a universal nanobody framework capable of accepting diverse CDRs
- antibodies/nanobodies
- design/developability
Apr 21, 2026
Catalytic residues are usually destabilizing
- biophysics/catalysis
- biophysics/stability
Apr 21, 2026
Compactness is positively correlated with unfolding cooperativity but negatively correlated with stability
- biophysics/folding
- biophysics/stability
Apr 21, 2026
Computational models are less designable than experimental structures
- evidence/design-validation
- design/inverse-folding
Apr 21, 2026
Confidence metrics for diffusion-based structure prediction methods can be improved with minimal changes to conditioning representations
- prediction/confidence
- inference/guidance
Apr 21, 2026
Confidence metrics from structure prediction correlate with accuracy even in the absence of coevolutionary data
- prediction/confidence
Apr 21, 2026
Conformational dynamics facilitate the emergence of catalytic activity in non-enzyme ligand-binding proteins
Apr 21, 2026
Conformational entropy can decrease or increase following ligand binding
- biophysics/interactions
- biophysics/dynamics
Apr 21, 2026
Conformational sampling by AlphaFold3-generation methods can be achieved by scaling the pair representation
- prediction/ensembles
- inference/conditioning
Apr 21, 2026
Conformational sampling can smoothen fitness landscapes
- evolution/mutation-effects
- biophysics/dynamics
Apr 21, 2026
Consensus sequence thermostabilities and activities are sensitive to breadth of sequences used for calculation
- evolution/ancestry
- design/enzymes
Apr 21, 2026
Consensus sequences are less thermostable than ancestrally reconstructed sequences
- evolution/ancestry
- biophysics/stability
Apr 21, 2026
Consensus sequences can show greater expression than extant sequences
- evolution/ancestry
- design/developability
Apr 21, 2026
Contrastive fine-tuning PLMs on inverse folding embeddings of experimental structures but not computational models improves downstream tasks
Apr 21, 2026
Contrastive learning of PLM embeddings on functional annotation improves variant effect prediction and homolog detection
Apr 21, 2026
Contrastive learning on whole structures leads to learning of distinct substructures
- model-analysis/representation-geometry
- antibodies/architecture
Apr 21, 2026
Cooling samples for cryogenic crystallography or cryo-EM narrows the conformational ensemble
- evidence/measurements
- biophysics/ensemble-analysis
Apr 21, 2026
Correlation between sequence log-likelihood and variant effect prediction performance breaks down as PLMs get larger
Apr 21, 2026
Coupling strength of two substitutions is correlated with backbone distance
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
Cryo-EM particle counting does not reliably provide true Boltzmann distributions of molecules
- evidence/measurements
- biophysics/ensemble-analysis
Apr 21, 2026
Crystal structures can have nonphysiological salt bridges
- evidence/measurements
- biophysics/interactions
Apr 21, 2026
D2O has a stabilizing effect on protein structure
- evidence/measurements
- biophysics/stability
Apr 21, 2026
DL models excel at finding pockets but not docking into them
- prediction/ligand-docking
Apr 21, 2026
DNA sequence entropy is strongly predictive of protein expression
- prediction/stability-expression
- design/developability
Apr 21, 2026
De novo designed proteins (from language models) and naturally occurring proteins borrow heavily from other functionally similar sequences
- evidence/generalization
- design/sequence-generation
Apr 21, 2026
De novo structure prediction with ligands outperforms apo de novo structure prediction alone
Apr 21, 2026
Deep learning methods produce different CDRH3 conformers
- prediction/antibody-structure
- prediction/ensembles
Apr 21, 2026
DiffDock confidence is inversely correlated with RMSD
- prediction/ligand-docking
- prediction/confidence
Apr 21, 2026
Different versions of low-rank adaptation have equivalent performance after controlling for learning rate
- training/fine-tuning
Apr 21, 2026
Dynamic time warping
- evolution/homology
- antibodies/architecture
Apr 21, 2026
Encoder- and decoder-based models have different strengths and are equally useful on generative tasks
- model-design/architectures
- model-design/generative-models
Apr 21, 2026
Eukaryotic viruses have more disordered proteins than prokaryotic viruses
- evolution/selection
- biophysics/dynamics
Apr 21, 2026
Fast dynamics of proteins drive slower conformational changes
- biophysics/dynamics
Apr 21, 2026
Fitness landscape navigability is correlated with its kurtosis and the centeredness of its maximum
- evolution/mutation-effects
- design/directed-evolution
Apr 21, 2026
Fitness prediction using inverse folding can be improved with auxiliary B-factor and SASA losses
Apr 21, 2026
Focused protein sequence libraries are poor training sets
- design/directed-evolution
- evidence/datasets
Apr 21, 2026
Functional proteins emerge from libraries of random sequences
- evolution/selection
- design/directed-evolution
Apr 21, 2026
GFP fitness data is best fit by simple models
- prediction/variant-effects
Apr 21, 2026
Gaussian processes can identify false positives
- evidence/measurements
Apr 21, 2026
Germline usage determines whether nanobody CDR3 is kinked or extended
Apr 21, 2026
Gibbs free energy
- biophysics/thermodynamics
Apr 21, 2026
Gibbs sampling is less effective than simpler algorithms for sequence design using PSSMs
Apr 21, 2026
Grafting catalytically important loops from one enzyme to another can transfer specificity and function
- design/enzymes
- biophysics/catalysis
Apr 21, 2026
Grokking
- training/pretraining-and-scaling
- training/fine-tuning
Apr 21, 2026
Guidance potentials can be added to diffusion-based structure prediction for enhanced sampling of protein conformations
- prediction/ensembles
- inference/guidance
Apr 21, 2026
Hallucination of CDRs
- design/antibodies
- inference/guidance
Apr 21, 2026
Helices can be stabilized by positively-charged residues in final turn
- biophysics/stability
Apr 21, 2026
High predicted aligned error correlates with violation of experimental distance measurements
- prediction/confidence
- evidence/measurements
Apr 21, 2026
High-abundance enzymes tend to be comprised of less costly amino acids
- evolution/selection
Apr 21, 2026
High-accuracy computational models might not always be effective for ligand docking
- prediction/ligand-docking
- evidence/design-validation
Apr 21, 2026
High-pLDDT designs can be insoluble
Apr 21, 2026
Including crystal contacts in training data improves sidechain prediction
- prediction/structure
- evidence/datasets
Apr 21, 2026
Including nonfunctional sequences during training improves prediction of negatives but not positives
- evidence/datasets
- prediction/variant-effects
Apr 21, 2026
Including optimal growth temperature during pre-training of PLMs improves prediction and design of thermostability
Apr 21, 2026
Incorporating ancestral sequences during PLM training improves performance in downstream tasks
Apr 21, 2026
Indels better tolerated at termini
- evolution/mutation-effects
Apr 21, 2026
Individual V, D, and J gene usage is highly uneven
- antibodies/repertoires
Apr 21, 2026
Information about backbone atoms is lost when rebuilding full-atom protein models from alpha carbon trace alone
- prediction/structure
- design/inverse-folding
Apr 21, 2026
Insertion mutations are associated with gain-of-function
- evolution/mutation-effects
- biophysics/interactions
Apr 21, 2026
Insertions and deletions are more detrimental to stability than substitutions
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
Insertions more tolerated than deletions
- evolution/mutation-effects
Apr 21, 2026
Inverse folding methods outperform sequence design by hallucination
- design/inverse-folding
- evidence/design-validation
Apr 21, 2026
Kinked nanobody CDR3s are longer than extended CDR3s
- antibodies/nanobodies
- antibodies/architecture
Apr 21, 2026
Layer normalization
- model-design/architectures
Apr 21, 2026
Learning rate decay
- training/objectives-and-optimization
Apr 21, 2026
Length independent clustering improves CDR prediction
- prediction/antibody-structure
- evolution/homology
Apr 21, 2026
Ligand-binding sites fit experimental electron density better than other parts of protein structures
- evidence/measurements
- prediction/ligand-docking
Apr 21, 2026
Likelihood of predicted contacts by MSA transformer weakly correlate with dwell time in molecular dynamics simulations
- model-analysis/interpretability
- biophysics/ensemble-analysis
Apr 21, 2026
Linear MD encoders outperform nonlinear encoders when extrapolating to unseen regions of protein conformational space
- biophysics/ensemble-analysis
- evidence/generalization
Apr 21, 2026
Lipid and detergent composition affects the conformational landscape of membrane proteins
- biophysics/dynamics
- evidence/measurements
Apr 21, 2026
Low-abundance proteins evolve faster
- evolution/selection
Apr 21, 2026
Low-rank Adaptation
- training/fine-tuning
Apr 21, 2026
Low-rank adaptation changes the magnitude, but not directionality, of matrix operations relative to fine-tuning
- training/fine-tuning
Apr 21, 2026
Low-rank adaptation with even fewer parameters can operate on truncated SVD matrices, but only with reinforcement learning
- training/fine-tuning
Apr 21, 2026
MD force fields can be screened for agreement with experimental data
- biophysics/molecular-simulation
- evidence/measurements
Apr 21, 2026
MD potentials from ML are more effective when protein-specific
- biophysics/molecular-simulation
Apr 21, 2026
MD simulations can distinguish enzyme design candidates for catalytically competent substates
Apr 21, 2026
ML-based protein design methods outperform physics-based methods at identifying deleterious sequences but not top sequences
- evidence/design-validation
- design/inverse-folding
Apr 21, 2026
Machine learning force fields for molecular dynamics require prior potentials
- biophysics/molecular-simulation
Apr 21, 2026
Markov State Models
- biophysics/ensemble-analysis
Apr 21, 2026
Masking ESMFold can sometimes sample alternate conformations
- prediction/ensembles
- inference/sampling-and-search
Apr 21, 2026
Mathematical terms in MD
- biophysics/thermodynamics
Apr 21, 2026
Mega-scale dataset is inaccurate for DNA-binding domains
- evidence/measurements
- evidence/datasets
Apr 21, 2026
Memorization in protein structure prediction
- evidence/generalization
- prediction/structure
Apr 21, 2026
Milk proteins are mostly negatively charged
Apr 21, 2026
Mixture-of-depths
- model-design/architectures
Apr 21, 2026
Models trained on synthetic data tend to converge to mean
- evidence/datasets
- evidence/generalization
Apr 21, 2026
Modifying conformational landscapes of proteins can give rise to new functions
- design/enzymes
- biophysics/dynamics
Apr 21, 2026
Molecular dynamics simulation of an entire cell
- biophysics/molecular-simulation
Apr 21, 2026
Most proteins have a single folding route
- biophysics/folding
Apr 21, 2026
Multiple nanobodies can bind simultaneously
- antibodies/nanobodies
- antibodies/recognition
Apr 21, 2026
Mutation of allosteric residues can influence the equilibrium dynamics of designed proteins
- biophysics/dynamics
- design/backbones
Apr 21, 2026
Mutations that give rise to new functions are not more destabilizing than mutations in general
- evolution/mutation-effects
- biophysics/stability
Apr 21, 2026
NPT ensemble
- biophysics/thermodynamics
- biophysics/molecular-simulation
Apr 21, 2026
Nanobodies are more easily expressed in bacteria and yeast than IgG antibodies
- antibodies/nanobodies
- design/developability
Apr 21, 2026
Natural selection favors the highest average local fitness
- evolution/selection
- evolution/mutation-effects
Apr 21, 2026
Natural sequences deviate from highest stability variants
- evolution/selection
- biophysics/stability
Apr 21, 2026
Neutral mutation frequency is similar across distance fitness peaks
- evolution/mutation-effects
Apr 21, 2026
No one-size-fits-all best approach to zero-shot or few-shot protein fitness prediction
- prediction/variant-effects
- evidence/design-validation
Apr 21, 2026
On fixed compute budgets, mixture-of-experts models outperform dense models
- model-design/architectures
- training/pretraining-and-scaling
Apr 21, 2026
Over-squashing
- model-design/architectures
Apr 21, 2026
Overtrained language models are more difficult to fine-tune
- training/fine-tuning
- training/pretraining-and-scaling
Apr 21, 2026
Paratope losses are required to enforce CDR-mediated antigen binding during de novo antibody design by hallucination
- design/antibodies
- inference/guidance
Apr 21, 2026
Poisoning attacks on LMs require a constant number of samples regardless of scale
- evidence/generalization
- evidence/datasets
Apr 21, 2026
Positional encoding
- model-design/architectures
Apr 21, 2026
Potts models and PSSMs can model fitness landscapes
- prediction/variant-effects
- evolution/mutation-effects
Apr 21, 2026
Protein denaturation cooperativity varies depending on whether it is thermally, chemically, or pressure induced
- biophysics/folding
- evidence/measurements
Apr 21, 2026
Protein dynamics can be used as filtering criteria for the design of highly selective small molecules
Apr 21, 2026
Protein structure is more evolutionarily conserved than dynamics
- evolution/selection
- biophysics/dynamics
Apr 21, 2026
ProteinGym
- evidence/datasets
- prediction/variant-effects
Apr 21, 2026
Proteins from halophilic organisms are more negatively charged
- evolution/selection
- design/developability
Apr 21, 2026
Psi torsion angles are effective metadynamic CVs for CDRH3 and CDRL3
Apr 21, 2026
Residual connections smoothen the loss landscape of neural networks
- model-design/architectures
- training/objectives-and-optimization
Apr 21, 2026
Rigidification of active site residues accelerates enzyme activity
Apr 21, 2026
Scaling hypothesis
- training/pretraining-and-scaling
Apr 21, 2026
Second-shell mutations can yield positive epistasis
- evolution/mutation-effects
- design/enzymes
Apr 21, 2026
Secondary structure losses are required to enforce CDR loopiness during de novo antibody design by hallucination
- design/antibodies
- inference/guidance
Apr 21, 2026
Sequence clusters and structural clusters of CDRH3 do not overlap
- antibodies/architecture
- evolution/homology
Apr 21, 2026
Sequence- and structure-derived ML quality metrics from ML do not correlate with each other
- evidence/design-validation
- prediction/confidence
Apr 21, 2026
Small energy barriers separate unfolded and folded states of small fast-folding proteins
- biophysics/folding
- biophysics/thermodynamics
Apr 21, 2026
Source of immunogenicity in VHH nanobodies
- design/developability
- antibodies/nanobodies
Apr 21, 2026
Spearman correlations of protein property prediction methods do not correlate perfectly with absolute error
- evidence/measurements
Apr 21, 2026
Structural models can be aligned at their ligand-binding sites using weighted-exponential term
- prediction/ligand-docking
- evolution/homology
Apr 21, 2026
The Evoformer can calculate ligand-binding residues
- prediction/binding
- model-analysis/interpretability
Apr 21, 2026
The latent space of VAEs can encode the conformational landscape of dynamic proteins
- biophysics/ensemble-analysis
- model-design/generative-models
Apr 21, 2026
The same protein can have two different structures across different organisms
- evolution/selection
- biophysics/dynamics
Apr 21, 2026
Transition matrices between metastable states approximate a protein's kinetics when their lag time exceeds relaxation time
- biophysics/ensemble-analysis
Apr 21, 2026
Viral protein MSAs fail to recapitulate contacts from evolutionary coupling analyses
- evolution/homology
- evidence/generalization
Apr 21, 2026
Weight-decomposed low-rank adaptation
- training/fine-tuning
Apr 21, 2026
Weighting training sequences by fitness improves identification of other fit sequences
- evidence/datasets
- prediction/variant-effects
Apr 21, 2026
Wildtype antibody frameworks can accommodate CDRs that bind to virtually any target
- design/antibodies
- antibodies/architecture
Apr 21, 2026
X-ray density can capture alternate conformations of proteins and their ligands
- evidence/measurements
- biophysics/ensemble-analysis
Apr 20, 2026
ALFA-tag
- design/modular-proteins
Apr 20, 2026
Affinity maturation
- antibodies/maturation
Apr 20, 2026
AlphaFold2
- prediction/structure
- model-design/architectures
Apr 20, 2026
AlphaFold3
Apr 20, 2026
Ancestral sequence reconstruction
- evolution/ancestry
Apr 20, 2026
Antibody developability
- design/developability
Apr 20, 2026
Antibody structure prediction
- prediction/antibody-structure
- prediction/complexes
Apr 20, 2026
B cells
- cell-biology/immune-signaling
- antibodies/maturation
Apr 20, 2026
BLOSUM62
- evolution/homology
- evolution/mutation-effects
Apr 20, 2026
Bayesian optimization
- design/directed-evolution
- inference/sampling-and-search
Apr 20, 2026
Chapman-Kolmogorov test
- biophysics/ensemble-analysis
Apr 20, 2026
Contrastive learning
Apr 20, 2026
Diffusion guidance
- inference/guidance
Apr 20, 2026
Disentangled attention
- model-design/architectures
Apr 20, 2026
Engineered trimerization domains
- design/modular-proteins
Apr 20, 2026
Focal loss
- training/objectives-and-optimization
Apr 20, 2026
Fourier transform
Apr 20, 2026
Hydrogen mass repartitioning
- biophysics/molecular-simulation
Apr 20, 2026
Immune repertoires
- antibodies/repertoires
Apr 20, 2026
Inverse folding
- design/inverse-folding
Apr 20, 2026
Light chains
- antibodies/architecture
Apr 20, 2026
Protein folding
- biophysics/folding
Apr 20, 2026
Protein structure tokenization
- model-design/multimodal
- evolution/homology
Apr 20, 2026
Residual connections
- model-design/architectures
Apr 20, 2026
RoseTTAFold
- prediction/structure
- model-design/architectures
Apr 20, 2026
Stability and thermostability
Apr 20, 2026
Uncertainty quantification
- prediction/confidence
Apr 20, 2026
Variant effect prediction
- prediction/variant-effects
Apr 20, 2026
pLDDT
- prediction/confidence
Apr 20, 2026
Binary classifiers
- evidence/measurements
Apr 20, 2026
Dimensionality reduction
- inference/feature-extraction
- model-analysis/representation-geometry
Apr 20, 2026
Fold-switching proteins
- biophysics/dynamics
- evolution/selection
Apr 20, 2026
Kd differs from IC50, LD50, and GI50
- evidence/measurements
- prediction/binding
Apr 20, 2026
Kemp eliminase
- design/enzymes
- biophysics/catalysis
Apr 20, 2026
Model merging
- training/fine-tuning
Apr 20, 2026
Time-lagged independent component analysis
- biophysics/ensemble-analysis
Apr 20, 2026
Evolution and natural selection
- evolution/selection
- evolution/mutation-effects
Apr 20, 2026
Alignment
- evolution/homology
Apr 20, 2026
Diffusion models
- model-design/generative-models
Apr 20, 2026
Epistasis
- evolution/mutation-effects
Apr 20, 2026
Helicase-assisted continuous editing
- design/directed-evolution
Apr 20, 2026
Protein backbone design
- design/backbones
Apr 20, 2026
Protein language models
- training/pretraining-and-scaling
- inference/feature-extraction
Apr 20, 2026
Structure prediction
- prediction/structure
Apr 20, 2026
Top2018
- evidence/datasets
- prediction/structure
Apr 17, 2026
Attention (machine learning)
- model-design/architectures
Apr 17, 2026
BERT
- model-design/architectures
- training/pretraining-and-scaling
Apr 17, 2026
Catastrophic forgetting
- training/fine-tuning
Apr 17, 2026
Chemokines
- cell-biology/immune-signaling
Apr 17, 2026
Clustering
Apr 17, 2026
Collective variables
- biophysics/ensemble-analysis
Apr 17, 2026
Decoy receptors
- cell-biology/immune-signaling
- cell-biology/therapeutic-mechanisms
Apr 17, 2026
E3 ubiquitin ligases
- cell-biology/therapeutic-mechanisms
Apr 17, 2026
Enhanced sampling
- biophysics/molecular-simulation
Apr 17, 2026
Graph neural networks
- model-design/architectures
Apr 17, 2026
Heterodimerization domains
- design/modular-proteins
- biophysics/interactions
Apr 17, 2026
Intrinsically disordered regions
- biophysics/dynamics
Apr 17, 2026
MD simulations
- biophysics/molecular-simulation
Apr 17, 2026
Major histocompatibility complex I
- cell-biology/immune-signaling
Apr 17, 2026
Metadynamics
- biophysics/molecular-simulation
Apr 17, 2026
Miscellaneous data notes
Apr 17, 2026
Mixture-of-experts
- model-design/architectures
Apr 17, 2026
Natural killer cells
- cell-biology/immune-signaling
Apr 17, 2026
Normal mode analysis
- biophysics/ensemble-analysis
Apr 17, 2026
OX40
- cell-biology/immune-signaling
- cell-biology/therapeutic-mechanisms
Apr 17, 2026
Potts models
- evolution/mutation-effects
Apr 17, 2026
Protein engineering
- design/modular-proteins
- design/directed-evolution
Apr 17, 2026
T cells
- cell-biology/immune-signaling
Apr 17, 2026
T-cell receptors
- cell-biology/immune-signaling
- antibodies/architecture
Apr 17, 2026
Tumor necrosis factors
- cell-biology/immune-signaling
Apr 17, 2026
Weight averaging
- training/objectives-and-optimization
Apr 10, 2026
Protein-protein interactions
- biophysics/interactions
Apr 10, 2026
Rosetta
- design/sequence-generation
- prediction/structure
Nov 04, 2024
Fréchet distance
- evidence/measurements
- evidence/design-validation
Jun 04, 2024
Transformers attend to gap tokens to "opt out" of updates
- model-analysis/interpretability
- model-design/architectures

Created with Quartz v4.5.2 © 2026

GitHub
Discord Community