Find open-source science resources

Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science with 140+ ready-to-use skills and 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Antigravity, and the open Agent Skills standard (K-Dense-AI, 26K+ stars, 2025)

PureJsImage is a free, open-source TypeScript library for decoding, inspecting, processing, and converting ordinary images and scientific rasters in Node.js and modern browsers. It provides explicit readers for microscopy, whole-slide pathology, medical imaging, electron microscopy, spectroscopy, hyperspectral, and multidimensional array formats. These include OME-TIFF, OME-Zarr, Aperio SVS, DICOM, NIfTI, MRC/CCP4, NRRD, DigitalMicrograph, EMD, ENVI, and FITS. Range-backed readers can request selected regions, tiles, volume planes, and metadata while preserving native numeric samples where supported. The default package has no runtime dependencies. Optional JPEG and PNG WebAssembly accelerators require explicit registration.

Translate differential transcript usage results into discrete splice events.

Modular Python suite for Neuro-AI research across all modalities, providing efficient data loaders (NeuralSet), curated datasets (NeuralFetch), scalable training (NeuralTrain), and unified benchmarking (NeuralBench) for building and evaluating neuroscience foundation models (Meta FAIR, 270+ stars, MIT License, 2026)

Neural network-based exchange-correlation functional for density functional theory (DFT) that surpasses state-of-the-art hybrid functionals in accuracy for main-group thermochemistry, kinetics, and non-covalent interactions at semi-local DFT cost; includes PySCF/GPU4PySCF/ASE bindings and C++/Fortran integrations (248+ stars, MIT License)

SpectralTAD is an R package designed to identify Topologically Associated Domains (TADs) from Hi-C contact matrices. It uses a modified version of spectral clustering that uses a sliding window to quickly detect TADs. The function works on a range of different formats of contact matrices and returns a bed file of TAD coordinates. The method does not require users to adjust any parameters to work and gives them control over the number of hierarchical levels to be returned.

The Zarr specification defines a format for chunked, compressed, N-dimensional arrays. It's design allows efficient access to subsets of the stored array, and supports both local and cloud storage systems. Rarr aims to implement this specification in R with minimal reliance on an external tools or libraries.

Research ecosystem for rigorous and trustworthy AI scientists — a protocol and skill bundle that makes autonomous research verifiable, crystallized, and observable through structured, machine-executable research artifacts and five agent skills for research management, compilation, verification, visualization, and publication (ARA-Labs, 447+ stars, MIT License, 2026)

Open-source PyMOL plugin for membrane-aware review of predicted, designed and experimental protein structures. Membrane Visual QC provides planar membrane-relative geometry, residue core/interface classification, hydropathy and ligand-context review, solvent-accessibility context, PDBTM/OPM orientation-source checks, and reproducible batch reporting. It is designed as a review assistant rather than a biological structure validator.

Neural differential equations in Julia

Multi-LLM consensus framework for automated cell type annotation in single-cell transcriptomics, integrating predictions from 10+ large language models with iterative discussion and uncertainty quantification to reduce single-model biases, achieving up to 95% accuracy without reference datasets; available as CRAN R package and PyPI Python package with Scanpy/Seurat integration (2025)

Interactive and hardware-agnostic SDK for laboratory automation, enabling programmatic control of liquid handlers, plate readers, and other lab instruments across multiple vendors; foundational infrastructure for self-driving laboratories and AI-driven experimental execution (447+ stars)

Provides functionality for producing geometric representations of protein and RNA structures, and biological interaction networks.

spoQC is a modular framework for multimodal quality control (QC) of imaging-based spatially resolved transcriptomics (SRT). It independently evaluates cell segmentation, imaging, and transcript data to identify high-quality regions (HQRs) across entire tissue sections. In addition, spoQC uses Markov random fields (MRFs) to incorporate spatial dependencies and generate spatially refined QC masks.

A library and command-line tool for building and analyzing complex homogeneous microkinetic models from quantum chemistry calculations, with support for quasi-harmonic thermochemistry, quantum tunnelling corrections, molecular symmetries and more.

This package is a wrapper of Integrative Genomics Viewer (IGV). It comprises an htmlwidget version of IGV. It can be used as a module in Shiny apps.

E(3)-equivariant neural network interatomic potentials achieving DFT accuracy with up to 1000× less training data than invariant models, foundational architecture behind MACE and Allegro (Harvard, MIT, Nature Communications 2022)

First unified Lean 4 framework for neural-network specification, execution, and verification; tensor shapes are part of the types, models are executable Lean programs, and the same definitions can be used by training code, graph transformations, certificate checkers, and proofs with CPU/CUDA backends (123+ stars, MIT License)

Co-create PowerPoint presentations with Generative AI from documents or topics

Exact, validated excision of coordinate-defined genomic regions from transposed NEXUS matrices.

Hand-curated Snakemake pipelines to combine identifier cross-references from multiple sources across dozens of biomedical types, including anatomical entities, diseases and phenotypes, genes and proteins and many others.

BRANCHSNV reports strict clade-exclusive nucleotide markers separately from single-nucleotide substitutions reconstructed on a selected edge of a rooted phylogenetic tree, while retaining ambiguity across equally parsimonious ancestral-state reconstructions.

edfcore is a zero-dependency TypeScript library for reading EDF, EDF+, BDF, and BDF+ physiological recordings in browser and Node.js applications. It provides programmatic access to biosignal samples, channel metadata, per-channel sampling rates, physical units, annotations, and discontinuous recording timelines.

A transparent, unit-aware calculator for the mathematical relationship between peptide mass, target concentration and solution volume. It normalizes mg, micrograms, mL and microlitres, shows the formula and includes a reference syringe visualization. Research-use-only software: it does not select a solvent, validate a laboratory method, calculate a dose or provide administration guidance.

A benchmark for ML-guided high-throughput materials discovery.

AI-assisted structural engineering workspace for AEC workflows: natural language to structural model, analysis, code-check, and report (171+ stars, MIT License, 2026)

Simple and accurate de novo protein binder design pipeline using AlphaFold2 backpropagation, MPNN, and PyRosetta for automated binder discovery (bioRxiv 2024)

Composite-objective protein design framework integrating Boltz, AlphaFold2, OpenFold3, ProteinMPNN, and ESM via JAX-based gradient optimization over continuous relaxed sequence space for multi-property binder design (319+ stars, MIT License, 2025)

Local Python sequence utilities for nucleotide composition, DNA and RNA reverse complements, NCBI genetic-code translation, six-frame candidate ORF enumeration, and IUPAC motif searches. Computase accepts raw nucleotide strings or one FASTA record and returns structured, bounded results with explicit scientific conventions.

Lightweight Markdown-only skills for autonomous ML research with cross-model review loops, idea discovery, and experiment automation; no framework lock-in, works with Claude Code, Codex, OpenClaw, or any LLM agent (12.8K+ stars, MIT License, 2026)

PathForge is a modular benchmarking framework for multiple instance learning in computational pathology. It supports whole slide image feature extraction, HDF5 artifact generation, tile overviews, benchmarking, pipeline optimization, classification, regression, survival and retrieval tasks, and support for model inference and visualization.

Biomedical Model Context Protocol (MCP) server unifying literature search across PubMed/Europe PMC, entity pivoting across genes/variants/drugs/diseases/pathways/proteins, local study analytics, and Claude Code/Codex integration for agentic biomedical research (531+ stars, MIT License, 2025-2026)

Open-source 3D architectural editor with an AI design assistant; build floor plans with walls, doors, windows, and furniture using natural language, with real-time WebGPU-powered previews (TangSY, 59+ stars, MIT License, 2026)

Beyond text-to-slides generation with PPTEval multi-dimensional evaluation (EMNLP 2025)

Deep learning library for Chemistry based on Tensorflow

ImageArray provides a framework for on-disk and in-memory image arrays, specifically for pyramidal images stored in HDF5, Zarr and life sciences image file formats (OME Bio-Formats).

A comprehensive toolkit that bridges popular Python-based immune repertoire analysis tools and Hugging Face protein language models into the R environment. Provides unified interfaces for TCR distance calculations (tcrdist3), sequence generation probability (OLGA), selection inference (soNNia), clustering (clusTCR), protein embeddings (ESM-2), metaclone discovery (metaclonotypist). Fully compatible with the scRepertoire and immApex ecosystem for single-cell immune repertoire analysis.

Shared multimodal AI agent layer for geospatial Python packages (leafmap, geoai, geemap, STAC, NASA Earthdata) and QGIS, exposing geospatial tools to LLMs with structured metadata, confirmation hooks, and support for OpenAI, Anthropic, Google Gemini, Ollama, and more; includes the OpenGeoAgent QGIS plugin (456+ stars, MIT License)

Library for fast calculations of **mo**lecula**r** **fe**at**u**re**s** from 3D structures for machine learning with a focus on steric descriptors.

Community-driven model zoo and deployment infrastructure for AI-powered bioimage analysis, enabling standardized sharing, validation, and cross-platform execution of deep learning models across Fiji, Ilastik, napari, and other scientific imaging tools (EPFL, EMBL, and global collaborators, actively maintained)

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

Converts Protein Data Bank structures into 3D-printable models. Each polymer chain is meshed separately and written as a named object in a single 3MF file, so a multi-material printer can assign one filament per chain. Protein chains can be rendered as a solvent-excluded surface, a cartoon, or a backbone tube; nucleic acids as a tube-and-rung form with the strands of a duplex welded at every base pair. Press-fit magnet pockets are optionally placed at chain interfaces, so a complex comes apart where its subunits actually meet. All meshes are checked for watertightness before export.

A flexible pipeline, built with Nextflow, for the complete analysis of bacterial genomes.

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.

Lineagespot is a framework written in R, and aims to identify SARS-CoV-2 related mutations based on a single (or a list) of variant(s) file(s) (i.e., variant calling format). The method can facilitate the detection of SARS-CoV-2 lineages in wastewater samples using next generation sequencing, and attempts to infer the potential distribution of the SARS-CoV-2 lineages.

ProSeqGO predicts Gene Ontology (GO) terms for protein sequences using ESM2 embeddings and a trained 1-Dimensional Convolutional Neural Network multi-label classifier. By integrating recent advances in protein language models, ProSeqGO facilitates large-scale, automated functional annotation directly from sequence input, empowering researchers to infer protein function, explore biological mechanisms, and accelerate discovery in genomics and proteomics.

Provides with toolkits to implement a full singIST analysis with pseudobulked Seurat objects of disease models and human data.

Python computational framework for analysis of single-molecule FRET data

Open-source LLM-powered R&D agent framework automating data-driven AI solution building through automated research, development, and evolution; achieves top open-source performance on MLE-Bench with dual Researcher-Developer agents and supports research copilot, data mining, Kaggle, and quant R&D workflows (13.6K+ stars, MIT License, 2025-2026)