Reproduce “ProtST: Multi-Modality Learning of Protein Sequences and Biomedical Texts” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Minghao Xu, Xinyu Yuan, Santiago Miret, Jian Tang
ProtST pairs protein sequences with biomedical descriptions during pretraining, aligning their representations while retaining sequence-only learning. Its dataset contains 553,052 aligned Swiss-Prot pairs covering four property fields. In supervised evaluation, ProtST-ESM-1b reached a mean fitness Spearman’s ρ of 0.772, versus 0.740 for ESM-1b across five tasks. Text-based zero-shot use requires annotations for descriptions and does not establish universal generalization.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.