Find open-source science resources

ECMWF's open-source machine-learning Earth system model developed by the WeatherGenerator Consortium with NVIDIA, trained on reanalyses, forecast data, and diverse observations across atmosphere, ocean, and land to provide a robust multi-scale model of Earth system dynamics; the first released version (v0.1, trained on ERA5) demonstrates global probabilistic forecasting skill on par with established AI models, with open training framework and config-driven multi-dataset ingestion pipeline (Apache 2.0)

An ontology based on PRO for describing the contributions that may be made, and the roles that may be held by a person with respect to a journal article or other publication (e.g. the role of article guarantor or illustrator).

An ontology for describing the steps in the workflow associated with the publication of a document or other publication entity.

UCell is a package for evaluating gene signatures in single-cell datasets. UCell signature scores, based on the Mann-Whitney U statistic, are robust to dataset size and heterogeneity, and their calculation demands less computing time and memory than other available methods, enabling the processing of large datasets in a few minutes even on machines with limited computing power. UCell can be applied to any single-cell data matrix, and includes functions to directly interact with SingleCellExperiment and Seurat objects.

Auto-generates clean, customizable academic CVs from open research data (OpenAlex, ORCID, Crossref, DataCite, Open Editors Plus). A single canonical CV object drives every output format (HTML, PDF, DOCX, LaTeX, Markdown); citations render through CSL; and the account holder is matched by persistent identifier (ORCID / OpenAlex ID) rather than name string. Free for individuals, open-source, and FAIR by design.

A seamless interface to the MEME Suite family of tools for motif analysis. 'memes' provides data aware utilities for using GRanges objects as entrypoints to motif analysis, data structures for examining & editing motif lists, and novel data visualizations. 'memes' functions and data structures are amenable to both base R and tidyverse workflows.

Python astronomy tools

Generalist autonomous research agent that grows a hypothesis tree to optimize any measurable task, beating Claude Code and Codex by 2.5× on the same compute budget across BrowseComp, Terminal-Bench 2.0, math reasoning, and MLE-Bench Lite; supports native CLI, keyless Claude Code/Codex integration, and an MCP tool server (RUC-NLPIR, 866+ stars, Apache 2.0, 2026)

Chemical reaction network and systems biology interface for scientific machine learning (SciML), enabling high-performance, GPU-parallelized simulation and analysis of complex biochemical systems with O(1) solvers (SciML, 518+ stars, Julia)

High-performance symbolic regression for discovering interpretable scientific equations from data, multi-population evolutionary search with Python/Julia backend, widely used in physics and astronomy (Cambridge, NeurIPS 2023)

Language agent gymnasium for challenging scientific tasks including DNA manipulation, literature search, and protein engineering

Deep learning package for many-body potential energy representation and molecular dynamics, achieving quantum-mechanical accuracy with classical MD efficiency (DeepModeling, Gordon Bell Prize 2020, 1.9k+ stars)

Hand-curated knowledge base of mTOR biology, linking 360+ primary studies to genes, drugs, diseases, pathways and outcomes. Every study carries a code for the kind of study behind it - synthesis of human data, human, animal, molecular or review - which names the system a finding was established in rather than ranking it, alongside open research questions.

Library of descriptors to aid in the data-mining of materials properties, created by the Lawrence Berkeley National Laboratory.

Curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines, enabling reproducible social science research with AI agents (Stanford REAP & CoPaper.AI, 3K+ stars, 2026)

With the DataPLANT biology ontology (DPBO), DataPLANT provides an intermediate ontology that acts as a broker and bridge between the individual researcher/domain experts and main ontology providers. DPBO enables easy and agile collection of missing vocabulary as well as relationships between terms for (meta)data annotation using DataPLANT’s Swate tool.

ramr is an R package for detection of epimutations (i.e., infrequent aberrant DNA methylation events) in large data sets obtained by methylation profiling using array or high-throughput methylation sequencing. In addition, package provides functions to visualize found aberrantly methylated regions (AMRs), to generate sets of all possible regions to be used as reference sets for enrichment analysis, and to generate biologically relevant test data sets for performance evaluation of AMR/DMR search algorithms.

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

The Simplified Upper Level Ontology (SULO) is ontology with a minimal set of classes and relations to guide the development of a personal health knowledge graph. [from homepage]

Democratizing AI scientists by transforming any LLM into research systems with 600+ scientific tools (Harvard MIMS)

A classification of subjects in Hochschule (universities of applied sciences)

A controlled vocbualry of higher education resource types

Fits second-order autoregressive AR(2) models to gene expression time series and reports the eigenvalue modulus |lambda|, a single statistic quantifying temporal persistence: how strongly a gene's recent past constrains its next value. Ranks genes into a clock/target/background hierarchy and reports correlation length, half-life and root type (real or complex) per gene.

Robust, lightweight infrastructure for multi-agent autonomous self-evolution, built for autoresearch; agents run in isolated git worktrees, share knowledge through a common state directory, and are scored by a grader daemon; natively integrated with Claude Code, Codex, Cursor Agent, OpenCode, and Kiro (672+ stars, Apache 2.0)

There are increasing demands on designing virus mutants with specific dinucleotide or codon composition. This tool can take both dinucleotide preference and/or codon usage bias into account while designing mutants. It is a powerful tool for in silico designs of DNA sequence mutants.

Predicts transcription factor binding sites in up to 316 vertebrate species by scoring JASPAR matrices against Ensembl promoter sequences and combining the match with seven contextual experimental datapoints, including evolutionary conservation, CAGE-defined transcription start sites, eQTLs, ChIP-seq peaks, ATAC-seq accessibility, DNase footprints and gene expression correlation, into a single score per site.

Principle-first scientific idea discovery framework that extracts reusable principles from public literature and private research materials, composes them into traceable Idea Cards with prior-art comparisons, and exports validation-ready research packs; emphasizes inspectable scientific objects, risk disclosure, and falsification paths (ICML 2026, 411+ stars, MIT License)

BgeeCall allows to generate present/absent gene expression calls without using an arbitrary cutoff like TPM<1. Calls are generated based on reference intergenic sequences. These sequences are generated based on expression of all RNA-Seq libraries of each species integrated in Bgee (https://bgee.org).

The EMI ontology is used to structure spectrum annotation provenance by reusing the PROV-O ontology (a W3C recommendation) and sample and observation data by applying the SOSA ontology. EMI reuses the SOSA ontology as a data schema for struturing the Sample and Observation data. SOSA (Sensor, Observation, Sample, and Actuator) is a subset of SSN (Semantic Sensor Network Ontology) that is a W3C recommendation. [from homepage]

Given a set of KEGG orthology group identifiers, identify the KEGG modules whose definition is satisfied

Benchmark evaluating AI agents on complex real-world scientific workflows in terminal environments across life, physical, earth, and mathematical sciences; featured on model cards for Claude Opus 4.7, GPT-5.5, and Gemini 3.1 Pro (200+ stars, Apache 2.0)

Official free and open-source PyTorch implementation of the ANI family of neural network interatomic potentials, shipping pre-trained ANI-1ccx/ANI-1x/ANI-2x ensembles with energy and force training, built-in unit conversion, and ASE integration; the foundational ANI ecosystem widely used for organic and biomolecular ML force field workflows (AIQMS, 560+ stars, MIT License, actively maintained)

Save Bioconductor data structures into file artifacts, and load them back into memory. This is a more robust and portable alternative to serialization of such objects into RDS files. Each artifact is associated with metadata for further interpretation; downstream applications can enrich this metadata with context-specific properties.

Scientific Computing for Chemists with Python is a Jupyter book teaching basic python in chemistry skills, including relevant libraries, and applies them to solving chemical problems.

Text-space optimizer that treats agent skill documents as trainable parameters for frozen LLMs, using scored rollouts and held-out validation gates to iteratively improve reusable natural-language skills; includes SkillOpt-Sleep for nightly self-evolution and improves accuracy across Claude Code, Codex, Copilot, and direct-chat harnesses, making it a meta-tool for evolving scientific agent skill workflows (15.5K+ stars, MIT License, PyPI)

A Simulation Tool for Fractured and Deformable Porous Media.

Visualization intermediate language that lets AI agents create expressive, polished charts from simple, human-editable specs, compiling the same input to 30+ chart types across Vega-Lite, ECharts, and Chart.js with an MCP server for agent integration (1.9K+ stars, MIT License, 2026)

TransformerEngine-accelerated checkpoints and training recipes for scaling biological foundation models (ESM-2, AMPLIFY, Geneformer, CodonFM) from single-GPU prototyping to multi-node FSDP training with FP8/MXFP8/NVFP4 precision, compatible with PyTorch, HF Accelerate, and PyTorch Lightning, plus sparse-autoencoder interpretability tools for biological foundation models (851+ stars, 2026)

High-performance ML research

Foundation AutoResearch Operating System: blueprint-driven runtime for orchestrating AI research workflows from idea generation and experiments to paper writing and peer review (OpenNSWM-Lab, 2.4K+ stars, 2026)

Flow-based generative model for atomistic protein binder design with test-time optimization, SOTA on binder benchmarks (ICLR 2026 Oral, NVIDIA)

MITE (Minimum Information about a Tailoring Enzyme) is a data repository and associated data standard designed to capture the reaction- and substrate-specificities of tailoring enzymes. Community-driven and fully expert-reviewed, it represents enzymatic reactions using reaction SMARTS and links to established resources such as UniProt, NCBI GenPept, Rhea, and MIBiG. MITE serves as a knowledgebase for enzyme and pathway annotation, in silico biosynthesis, and machine learning applications.

netZooR unifies the implementations of several Network Zoo methods (netzoo, netzoo.github.io) into a single package by creating interfaces between network inference and network analysis methods. Currently, the package has 3 methods for network inference including PANDA and its optimized implementation OTTER (network reconstruction using mutliple lines of biological evidence), LIONESS (single-sample network inference), and EGRET (genotype-specific networks). Network analysis methods include CONDOR (community detection), ALPACA (differential community detection), CRANE (significance estimation of differential modules), MONSTER (estimation of network transition states). In addition, YARN allows to process gene expresssion data for tissue-specific analyses and SAMBAR infers missing mutation data based on pathway information.

NVIDIA and King's College London's open-source AI toolkit for healthcare imaging, providing foundational frameworks for medical image annotation (MONAI Label), training (MONAI Core), and deployment (MONAI Deploy) across radiology, pathology, and endoscopy (8K+ stars, Apache 2.0)

Incremental knowledge graph construction using LLMs with entity extraction and Neo4j visualization

BiocCheck guides maintainers through Bioconductor best practicies. It runs Bioconductor-specific package checks by searching through package code, examples, and vignettes. Maintainers are required to address all errors, warnings, and most notes produced.

Eukaryotic Genome Annotation Pipeline-External caller scripts and documentation

ReviewAid is an open-source AI-assisted tool for full-text screening and data extraction in systematic reviews. It supports evidence synthesis workflows by using large language models to classify articles according to user-defined PICO criteria and extract structured information from full-text publications. ReviewAid is designed as a supplementary reviewer rather than a replacement for human judgement. It aims to reduce manual workload, improve consistency, and assist researchers during screening and data extraction while maintaining human oversight throughout the evidence synthesis process.

markeR is an R package that provides a modular and extensible framework for the systematic evaluation of gene sets as phenotypic markers using transcriptomic data. The package is designed to support both quantitative analyses and visual exploration of gene set behaviour across experimental and clinical phenotypes. It implements multiple methods, including score-based and enrichment approaches, and also allows the exploration of expression behaviour of individual genes. In addition, users can assess the similarity of their own gene sets against established collections (e.g., those from MSigDB), facilitating biological interpretation.

GlycoDash is an R Shiny dashboard for processing glycomics data obtained from LaCyTools, SweetSuite and Skyline.