Quartz 4

❯

Clustering

Created Apr 10, 2026Modified Apr 17, 2026

Clustering refers to the process of grouping together data points. It also refers to a set of algorithms that carries this out.

Alternate conformations can be sampled with MSA-based structure prediction methods using custom PDB databases and subsampled MSAs
Alternate sequence clustering schemes outperform uniform sampling when training protein language models
BERT models trained on sequence clusters outperform those trained on all data
CDR representations segregate into distinct clusters
CUMAb
Conformational entropy in antibodies decreases during affinity maturation
Conformationally stabilizing mutations can be identified from sequence clusters that are predicted to fold into that conformation when passed to AlphaFold2
Diffusion-based structure prediction can sometimes model conformational ensembles
Dimensionality reduction
Length independent clustering improves CDR prediction
Markov State Models
PLM embeddings contain enough information to align proteins without fine-tuning
Protein structure more conserved than sequence
Sequence clusters and structural clusters of CDRH3 do not overlap