Find open-source science resources

ProtSent-V2 35M plus one more contrastive pass on a

protein-protein interaction (PPI) sites, designed to handle both structured

protein embeddings where biological similarity maps to embedding proximity. Intended

proteins and protein-ligand complexes based in the 3D structure

DupyliCate is a high-throughput Python tool designed for mining

A local command-line tool for ancestral sequence reconstruction with

Trinity-Mini-AI-Scientist

protein engineering, and single-cell omics analysis (PharMolix & Tsinghua

SMBGC Annotation using Neural Networks Trained on Interpro Signatures

Curated, multilingual library of 182 installable AI agent skills for

SMILES structures based on both disease pathways and protein sequence.

protein–protein and peptide–MHC interaction, and more — following the…

Py-HLA-Match is a Python library for standardised, rule

Autonomous AI agent for end-to-end spatial proteomics analysis

protein structure prediction and design that defines a new frontier

SMILES structures based on both disease pathways and protein sequence.

protein-language-model-informed token space; enables zero-shot cell

650M with LoRA for predicting protein subcellular localization (10 classes).

Long-context generative genomic foundation model using 6-mer tokenization

Democratizing AlphaFold3: PyTorch reimplementation to accelerate protein structure prediction research

proteins, nucleic acids, small molecules, and cyclic peptides with fine

proteins, small molecules, DNA, RNA, glycans) achieving SOTA across benchmarks

Gravity-bio-16B-A3B is a biology-focus midtrained model

proteins, antibodies, small molecules, chemical reactions, materials, and their spatial

Molexar-10M Base is the unconditional base model for Molexar

Molexar-10M Omni is the universal multi-condition model for

protein target sequences. The mirnaprotpred package provides two core modules

both molecules and proteins within a single decoder-only architecture.

promptable and aligned protein language models for protein sequence generation.

KAU-BioMedLLM is a research prototype for source-grounded biomedical

Cross-platform system optimizations for accelerating AlphaFold3 training with 1.73

Learning the language of protein-protein interactions

protein structure determination from electron density with high accuracy (3DEM/EMBL

protein from a single anchor input, with optional sequence-specific

Access to Biological Web Services from Python.

AMPSeek is a workflow designed to predict the antimicrobial activity

Computes weighed laboratory buffer recipes from target pH, concentration, and

protein sequences, 236 million protein structures, and 539 million proteins

This model card provides an overview of the intended use

This model card provides an overview of the intended use

This set of model weights was released with the GitHub

protein language model for protein sequence generation. It is the smallest

protein language model for protein sequence generation. It is the largest

proteins via backpropagation through compositional design constraints parameterized by structure

!Protein-ligand interaction header

First fully open-source model achieving AlphaFold3-level accuracy with

Decentralized self-organizing teams of AI agents for long-running

This set of model weights was released with the GitHub

CGAS is an automated Python-based pipeline providing a comprehensive

acid and protein language, integrating DNA/RNA/protein sequences (Nature Machine Intelligence 2025)