Find open-source science resources

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)

Imputes HLA classical alleles using GWAS SNP data, and it relies on a training set of HLA and SNP genotypes. HIBAG can be used by researchers with published parameter estimates instead of requiring access to large training sample datasets. It combines the concepts of attribute bagging, an ensemble classifier method, with haplotype inference for SNPs and HLA types. Attribute bagging is a technique which improves the accuracy and stability of classifier ensembles using bootstrap aggregating and random variable selection.

Ontologies that aim to provide semantic specifications for units of measure, quantity kind, dimensions and data types.

The modern C++ library for sequence analysis.

Genome mapping and spliced alignment of cDNA or amino acid sequences

A teaching platform for computer-aided drug design (CADD) using open source packages and data.

Agricultural machine learning platform

Provides a comprehensive suite of functions to design and annotate CRISPR guide RNA (gRNAs) sequences. This includes on- and off-target search, on-target efficiency scoring, off-target scoring, full gene and TSS contextual annotations, and SNP annotation (human only). It currently support five types of CRISPR modalities (modes of perturbations): CRISPR knockout, CRISPR activation, CRISPR inhibition, CRISPR base editing, and CRISPR knockdown. All types of CRISPR nucleases are supported, including DNA- and RNA-target nucleases such as Cas9, Cas12a, and Cas13d. All types of base editors are also supported. gRNA design can be performed on reference genomes, transcriptomes, and custom DNA and RNA sequences. Both unpaired and paired gRNA designs are enabled.

An ontology about scholarship

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.

Free-text promptable universal 3D medical image segmentation foundation model enabling zero-shot segmentation of diverse anatomical structures and pathologies via natural language prompts across CT, MRI, and other volumetric imaging modalities (DKFZ, 195+ stars, Apache 2.0)

Rapid & standardized annotation of bacterial genomes, MAGs & plasmids

This package has for objectives to provide a method to make Linear Models for high-dimensional designed data. limpca applies a GLM (General Linear Model) version of ASCA and APCA to analyse multivariate sample profiles generated by an experimental design. ASCA/APCA provide powerful visualization tools for multivariate structures in the space of each effect of the statistical model linked to the experimental design and contrarily to MANOVA, it can deal with mutlivariate datasets having more variables than observations. This method can handle unbalanced design.

Deep learning software to decode EEG, ECG or MEG signals, providing standardized neural network models, preprocessing pipelines, and evaluation workflows for brain-computer interfaces and cognitive neuroscience research (1.2K+ stars, BSD 3-Clause, actively maintained)

SOTA multimodal document parsing with 1.2B parameters outperforming GPT-4o, converts PDFs to LLM-ready Markdown/JSON

This repository contains a 4-bit GPTQ quantized version of google/medgemma-27b-text-it, optimized for high-throughput inference using vLLM and the Marlin kernel.

The understanding of cancer mechanism requires the identification of genes playing a role in the development of the pathology and the characterization of their role (notably oncogenes and tumor suppressors). We present an updated version of the R/bioconductor package called MoonlightR, namely Moonlight2R, which returns a list of candidate driver genes for specific cancer types on the basis of omics data integration. The Moonlight framework contains a primary layer where gene expression data and information about biological processes are integrated to predict genes called oncogenic mediators, divided into putative tumor suppressors and putative oncogenes. This is done through functional enrichment analyses, gene regulatory networks and upstream regulator analyses to score the importance of well-known biological processes with respect to the studied cancer type. By evaluating the effect of the oncogenic mediators on biological processes or through random forests, the primary layer predicts two putative roles for the oncogenic mediators: i) tumor suppressor genes (TSGs) and ii) oncogenes (OCGs). As gene expression data alone is not enough to explain the deregulation of the genes, a second layer of evidence is needed. We have automated the integration of a secondary mutational layer through new functionalities in Moonlight2R. These functionalities analyze mutations in the cancer cohort and classifies these into driver and passenger mutations using the driver mutation prediction tool, CScape-somatic. Those oncogenic mediators with at least one driver mutation are retained as the driver genes. As a consequence, this methodology does not only identify genes playing a dual role (e.g. TSG in one cancer type and OCG in another) but also helps in elucidating the biological processes underlying their specific roles. In particular, Moonlight2R can be used to discover OCGs and TSGs in the same cancer type. This may for instance help in answering the question whether some genes change role between early stages (I, II) and late stages (III, IV). In the future, this analysis could be useful to determine the causes of different resistances to chemotherapeutic treatments. An additional mechanistic layer evaluates if there are mutations affecting the protein stability of the transcription factors (TFs) of the TSGs and OCGs, as that may have an effect on the expression of the genes.

# DrugGen 2: A disease-aware language model for enhancing drug discovery DrugGen-2 is a disease‑aware language model specialized for generating drug-like SMILES structures based on both disease pathways and protein sequence.

CENO-P-1B is the multi-species alignment (MSA) post-trained variant of the 1B CENO DNA foundation model, for variant effect prediction (VEP). It carries intraencodingpattern in its config and ships the MSA scoring path (modelingcenop.py), which consumes a per-token seq_idx to score packed MSA…

CENO-1B-131k is the long-context (131k) checkpoint of the 1B CENO DNA foundation model — a causal language model over genomic sequence built on a Nemotron-H Mamba / Attention / Mixture-of-Experts hybrid backbone (no MSA inputs).

CENO-80M-1m is the long-context (1M) checkpoint of the 80M CENO DNA foundation model — a causal language model over genomic sequence built on a Nemotron-H Mamba / Attention / Mixture-of-Experts hybrid backbone (no MSA inputs).

Plain-text, git-tracked electronic lab notebook (ELN) for reproducible bioinformatics — threads your R & Python figures into living lab notes with full provenance. Built for single-cell / CyTOF / flow cytometry; works with Obsidian, Quarto & Jupyter.

Free, open-source desktop AI research assistant that runs locally and turns natural-language requests into real data analysis, literature search, figure generation, and manuscript review; ships with 149 scientific skills, 326 workflow templates, and 229 databases across genomics, proteomics, drug discovery, and materials science, plus a living lab notebook, 60+ scientific file previews, and LaTeX editing (K-Dense-AI, 908+ stars, MIT License, 2026)

Spatial transcriptomic technologies have helped to resolve the connection between gene expression and the 2D orientation of tissues relative to each other. However, the limited single-cell resolution makes it difficult to highlight the most important molecular interactions in these tissues. SpaceMarkers, R/Bioconductor software, can help to find molecular interactions, by identifying genes associated with latent space interactions in spatial transcriptomics.

World's first fully open, accelerated weather AI software stack with Medium Range forecasting and Nowcasting models using generative AI (January 2026)

An R Package for Geneset Enrichment Workflows.

Fast, interactive, multi-dimensional image viewer for Python, foundational platform for scientific imaging AI with a rich plugin ecosystem integrating deep learning segmentation, object tracking, and microscopy analysis workflows (2.6K+ stars)

High-performance, GPU-accelerated library for key computational chemistry tasks including molecular similarity, conformer generation, and geometry relaxation, designed to accelerate drug-discovery and molecular-modeling workflows (264+ stars, Apache 2.0)

An interface to the fast-access storage format for VCF data provided in SeqArray, with tools for common operations and analysis.

An issue on the UBERON GitHub Issue tracker

MagNET is a family of neural networks for predicting NMR chemical shifts. This repository contains all models, datasets, and code to reproduce the data in the MagNET paper.

Neural network-based cryo-EM heterogeneous reconstruction, modeling continuous 3D structure distributions from single-particle images, with CryoDRGN-ET extending to in-cell cryo-electron tomography (MIT CSAIL, Nature Methods 2021/2024)

Universal machine learning interatomic potential for atomistic simulation of materials, molecules, and biomolecules across the periodic table, with open-source pretrained models and inference tools (Orbital Materials, 2024-2025)

multiHiCcompare provides functions for joint normalization and difference detection in multiple Hi-C datasets. This extension of the original HiCcompare package now allows for Hi-C experiments with more than 2 groups and multiple samples per group. multiHiCcompare operates on processed Hi-C data in the form of sparse upper triangular matrices. It accepts four column (chromosome, region1, region2, IF) tab-separated text files storing chromatin interaction matrices. multiHiCcompare provides cyclic loess and fast loess (fastlo) methods adapted to jointly normalizing Hi-C data. Additionally, it provides a general linear model (GLM) framework adapting the edgeR package to detect differences in Hi-C data in a distance dependent manner.

Open-source, all-atom biomolecular foundation model that turns co-folding into a scalable engine for structure prediction, design, and optimization across proteins, nucleic acids, and small molecules in drug discovery; ranked first on PXMeter-AB, FoldBench-AB, and 2026ARK-AB antibody-antigen benchmarks (263+ stars, Apache 2.0)

💻 Github | 📄 Report (Coming soon...) | 🚀 Demo

!image

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)

Bioconductor-native infrastructure for handling large nanoporetech modkit bedMethyl pileup files from ONT data using HDF5Array and DelayedArray.

Pythonic Access to the Ensembl database.

Diffusion MR imaging.