Find open-source science resources

Benchmark quantifying end-to-end autonomous AI research abilities of LLM agents across 20 tasks from SOTA machine learning papers spanning NLP, code, math, biochemical modelling, and time series forecasting, with normalized score metrics against human SOTA and HuggingFace dataset

First physics-aligned interactive benchmark for LLM agents in engineering construction, designing rockets/cars/bridges in physics simulator with 3D spatial geometry library

Benchmark evaluating AI agents on 75 curated Kaggle-style ML engineering competitions with reproducible Docker-based grading harness, human baselines, and end-to-end task lifecycle, used as a primary benchmark for autonomous ML research agents (e.g., InternAgent #1 at 36.44%)

Unified latent diffusion transformer that jointly generates periodic crystals and non-periodic molecules, scaling to 500M parameters with SOTA results on QM9, MP20, and GEOM-DRUGS (Meta FAIR, ICML 2025, 310+ stars)

Accessible protein design platform via Google Colab integrating AlphaFold2, RoseTTAFold, and ProteinMPNN for de novo hallucination, fixed backbone design, and binder design (Sergey Ovchinnikov, 2022+)

Baidu's open-source reproduction of AlphaFold3 in PaddlePaddle, providing pretrained weights and inference pipelines for unified biomolecular structure prediction across proteins, nucleic acids, ligands, ions, and post-translational modifications within the PaddleHelix biocomputing platform (Baidu, bioRxiv 2024)

Pythonic access to FASTA files.

Multimodal deep learning framework integrating peptide-MHC protein sequence, structure, and biochemical properties to predict class-I immunogenicity for infectious disease epitopes and cancer neoepitopes with cancer-wildtype contrastive learning, enabling personalized vaccine design (Krishnaswamy Lab, Yale University)

GenBio AI's software stack for the AI-Driven Digital Organism, supporting adaptation and finetuning of multiscale biological foundation models across DNA, RNA, protein, structure, and single-cell tasks with reproducible CLIs and pretrained model zoo (2025)

Universal pretrained neural network potential with charge and magnetic moment awareness, trained on 1.5M+ Materials Project inorganic structures for charge-informed molecular dynamics and phase diagram prediction (Berkeley, Nature Machine Intelligence 2023 Cover)

Euclidean neural networks for arbitrary point transformations enabling E(3)-equivariant deep learning, foundational library for building geometry-aware neural networks in molecular dynamics, materials science, and physics

Self-supervised vision foundation model for generalized structural brain MRI analysis, pretrained on ~49,000 scans from diverse datasets and generalizing across brain age prediction, dementia/MCI classification, IDH mutation detection, glioma survival prediction, time-to-stroke estimation, MR sequence classification, and brain tumor segmentation; outperforms task-specific models especially with limited training data (Mass General Brigham & Harvard Medical School, 129+ stars)

Lightweight supervised slide foundation model with 0.9M parameters pretrained on 24K whole-slide images for pan-cancer morphological classification, achieving competitive performance with much larger self-supervised models (TITAN, GigaPath) while enabling finetuning on consumer-grade GPUs; includes standardized MIL implementations and benchmarking across 15+ classification tasks (Mahmood Lab, Harvard Medical School, 153+ stars)

Official implementation of the second-generation fully autonomous scientific discovery system, extending the original with agentic tree search and reduced template dependency to achieve workshop-level accepted papers (6.7K+ stars, 2025)

Graph neural network operating entirely at the atomic level for protein-ligand conformational ensemble prediction and docking, generating diverse solutions through rapid stochastic denoising to model conformational heterogeneity (Baker Lab, bioRxiv 2025)

Conversational data analysis using natural language

Cheminformatic extension for the SQLAlchemy database.

NIST's open-source platform for data-driven atomistic materials design, integrating DFT datasets (JARVIS-DFT), machine learning property prediction (JARVIS-ML), and a comprehensive leaderboard for benchmarking materials AI methods across the periodic table (384+ stars)

Large-scale flow-based protein backbone generator utilizing hierarchical fold class labels for conditioning with a tailored scalable transformer architecture, enabling controllable de novo protein design (264+ stars)

In silico directed evolution framework using few-shot active learning to optimize protein activities, enabling rapid protein engineering with minimal experimental data (352+ stars, 2023)

Systematic medical RAG toolkit for question answering over PubMed, StatPearls, textbooks, and Wikipedia, supporting multiple retrievers, domain LLMs, and follow-up-query workflows for benchmarked clinical/biomedical QA (ACL Findings 2024)

Vision-language pathology foundation model using contrastive learning on histopathology image-text pairs, enabling zero-shot classification, slide-level retrieval, and multimodal reasoning across diverse cancer types (Mahmood Lab, 494+ stars)

Python wrapper for [bedtools](https://github.com/arq5x/bedtools).

Descriptor computation(chemistry) and (optional) storage for machine learning.

Universal chart comprehension and reasoning model

Circlator is a tool to circularize genome assemblies. It will attempt to identify each circular sequence and output a linearised version of it. It does this by assembling all reads that map to contig ends and comparing the resulting contigs with the input assembly.

k-mer counting, filtering, and graph traversal.

A VCF Parser for Python.

Learning nonlinear operators

AI for chemical reaction prediction and synthesis planning

The Reagent Ontology (ReO) adheres to OBO Foundry principles (obofoundry.org) to model the domain of biomedical research reagents, considered broadly to include materials applied “chemically” in scientific techniques to facilitate generation of data and research materials. ReO is a modular ontology that re-uses existing ontologies to facilitate cross-domain interoperability. It consists of reagents and their properties, linking diverse biological and experimental entities to which they are related. ReO supports community use cases by providing a flexible, extensible, and deeply integrated framework that can be adapted and extended with more specific modeling to meet application needs.