PaTz β€” Physics-Aware Transformer with eXplainability
● API Offlinepatz_env

Data Exploration

2,620 PROTEINS

Dataset statistics for HIV-TB co-infection protein corpus. UniProtKB + PDB annotations.

Train / Val / Test Split
Loading chart…
Proteins by Organism
Loading chart…
Data Collection Summary
Total proteins
2620
Mean seq length
317.6 aa
With 3D structure (PDB)
841
Unique GO terms
354
MMseqs2 threshold
30% sequence identity
Source databases
UniProtKB, PDB, Gene Ontology