Find open-source science resources

GCAT is an association test for genome wide association studies that controls for population structure under a general class of trait models. This test conditions on the trait, which makes it immune to confounding by unmodeled environmental factors. Population structure is modeled via logistic factors, which are estimated using the `lfa` package.

MathModDB is a database of mathematical models developed by the Mathematical Research Data Initiative (MaRDI). MathModDB defines a data model with classes (Mathematical Model, Mathematical Formulation, Research Field, Research Problem, Quantity [Kind], Computational Task, Publication), object properties/relations, data properties and annotation properties as an ontology. This ontology is populated with individuals/data from various fields of applied mathematics, making it a knowledge graph. [from homepage]

Scientific equation discovery with agentic AI, elevating LLMs from equation proposers to autonomous scientists that write code, analyze data, implement equations, and optimize based on experimental feedback; outperforms baselines by 6-35% across four science disciplines with robustness to noise and out-of-domain generalization (GAIR-NLP / SJTU, 49+ stars, Apache 2.0)

WiNGPT 是一个基于GPT的医疗垂直领域大模型，旨在将专业的医学知识、医疗信息、数据融会贯通，为医疗行业提供智能化的医疗问答、诊断支持和医学知识等信息服务，提高诊疗效率和医疗服务质量。

Epialleles are specific DNA methylation patterns that are mitotically and/or meiotically inherited. This package calls and reports cytosine methylation as well as frequencies of hypermethylated epialleles at the level of genomic regions or individual cytosines in next-generation sequencing data using binary alignment map (BAM) files as an input. Among other things, this package can also extract and visualise methylation patterns and assess allele specificity of methylation.

This package runs the GADGETS method to identify epistatic effects in nuclear family studies. It also provides functions for permutation-based inference and graphical visualization of the results.

This package allows interactive viewing of package maintainer information. The Bioconductor Package Maintainer Application sends yearly verification emails to accept Bioconductor policies; this application also depicts maintainer status on opting in and if the email is deemed valid.

Matěj Pekár, Vít Musil, Rudolf Nenutil, Petr Holub, Tomáš Brázdil

scFeatures constructs multi-view representations of single-cell and spatial data. scFeatures is a tool that generates multi-view representations of single-cell and spatial data through the construction of a total of 17 feature types. These features can then be used for a variety of analyses using other software in Biocondutor.

Foundation model for joint segmentation, detection, and recognition of biomedical objects across nine imaging modalities, with v2 introducing BoltzFormer architecture for end-to-end 3D inference (Microsoft, Nature Methods 2025)

DNA sequence analysis

A generic but comprehensive bacterial annotation pipeline, built with Nextflow, with nice graphical options for investigating results.

A Shiny application to explore the TCGA Diagnostic Image Database.

This model is a fine-tuned version of google/medgemma-1.5-4b-it specialized for mammogram analysis and breast imaging interpretation.

Computes Multiple Co-Inertia Analysis (MCIA), a dimensionality reduction (jDR) algorithm, for a multi-block dataset using a modification to the Nonlinear Iterative Partial Least Squares method (NIPALS) proposed in (Hanafi et. al, 2010). Allows multiple options for row- and table-level preprocessing, and speeds up computation of variance explained. Vignettes detail application to bulk- and single cell- multi-omics studies.

Open-source toolkit and benchmark for learning-based theorem proving in Lean, providing programmatic Lean interaction, a 98K+ theorem dataset extracted from 217 Lean projects, and ReProver—the first retrieval-augmented LLM-based theorem prover for Lean—with reproducible training pipelines underpinning much subsequent Lean prover research (Caltech & NVIDIA, NeurIPS 2023 Outstanding Paper, Datasets & Benchmarks)

AVAILABLE NOW THE LATEST ITERATION OF THE ALOE FAMILY! ALOE BETA 8B AND ALOE BETA 70B VERSIONS. These include: Better overall performance More thorough alignment and safety * License compatible with more uses

Aloe: A Family of Fine-tuned Open Healthcare LLMs

Package performs summarization of replicates, filtering by frequency, several different options for imputing missing data, and a variety of options for transforming, batch correcting, and normalizing data.

The package comprises a set of pretrained machine learning models to predict basic immune cell types. This enables all users to quickly get a first annotation of the cell types present in their dataset without requiring prior knowledge. scAnnotatR also allows users to train their own models to predict new cell types based on specific research needs.

SCENIC+ is a python package to build gene regulatory networks (GRNs) using combined or separate single-cell gene expression (scRNA-seq) and single-cell chromatin accessibility (scATAC-seq) data.

This package contains functionality to run differential gene co-expression across two different conditions. The algorithm is inspired by Voigt et al. 2017 and finds Conserved, Specific and Differentiated genes (hence the name CSD). This package include efficient and variance calculation by bootstrapping and Welford's algorithm.

Design primers for targeted single-cell RNA-seq used by TAP-seq. Create sequence templates for target gene panels and design gene-specific primers using Primer3. Potential off-targets can be estimated with BLAST. Requires working installations of Primer3 and BLASTn.

Discrete diffusion framework for generative protein sequence design over evolutionary-scale databases, supporting unconditional generation, evolutionary-guided conditional design, motif scaffolding, and intrinsically disordered region generation through order-agnostic autoregressive diffusion, enabling sequence-only protein design without structural priors (Microsoft Research, Nature Communications 2024)

CebraEM is a bioinformatics tool for analyzing and processing large-scale imaging data, providing a pipeline for segmentation, annotation, and analysis with support for both Linux and Windows environments. It includes modules for core functionality, annotation, and network analysis, requiring specific dependencies and a conda environment for execution.

The package aims to identify miRNA sponge or ceRNA modules in heterogeneous data. It provides several functions to study miRNA sponge modules at single-sample and multi-sample levels, including popular methods for inferring gene modules (candidate miRNA sponge or ceRNA modules), and two functions to identify miRNA sponge modules at single-sample and multi-sample levels, as well as several functions to conduct modular analysis of miRNA sponge modules.

an automated workflow for the generation and storage of DFT calculations for organic molecules.

ICML 2025 drug discovery generalist using masked discrete diffusion and fragment-based generation with molecular context guidance (NVIDIA)

While large language models (LLMs) have achieved impressive progress, their application in scientific domains such as chemistry remains hindered by shallow domain understanding and limited reasoning capabilities. In this work, we focus on the specific field of chemistry and develop a Chemical…

ChemDFM-v2.0 is the latest non-thinking model of ChemDFM, the pioneering open-sourced dialogue foundation model for Chemistry and molecule science.

Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants.

Open-source, platform-independent, community-supported software for describing and comparing microbial communities

PII Detection Model | 44M Parameters | Open Source

PII Detection Model | 434M Parameters | Open Source

DeepMind's Olympiad-level geometry theorem prover combining neural language model with symbolic deduction engine, AlphaGeometry2 solves 84% of IMO geometry problems (42/50) at gold-medalist level (Nature 2024)

Standard data-centric AI package for data quality and machine learning, automatically detecting label errors, outliers, and dataset issues to improve scientific dataset reliability and model performance (11K+ stars, MIT License)

An efficient machine-learning interatomic potential for molecular dynamics simulations of organic and biological systems trained on an extension of the SPICE2 dataset. The model comes in two sizes (S and M).

SpotClean is a computational method to adjust for spot swapping in spatial transcriptomics data. Recent spatial transcriptomics experiments utilize slides containing thousands of spots with spot-specific barcodes that bind mRNA. Ideally, unique molecular identifiers at a spot measure spot-specific expression, but this is often not the case due to bleed from nearby spots, an artifact we refer to as spot swapping. SpotClean is able to estimate the contamination rate in observed data and decontaminate the spot swapping effect, thus increase the sensitivity and precision of downstream analyses.

TEKRABber is made to provide a user-friendly pipeline for comparing orthologs and transposable elements (TEs) between two species. It considers the orthology confidence between two species from BioMart to normalize expression counts and detect differentially expressed orthologs/TEs. Then it provides one to one correlation analysis for desired orthologs and TEs. There is also an app function to have a first insight on the result. Users can prepare orthologs/TEs RNA-seq expression data by their own preference to run TEKRABber following the data structure mentioned in the vignettes.

a specification for describing analysis workflows and tools that are portable and scalable across a variety of software and hardware environments, from workstations to cluster, cloud, and high performance computing (HPC) environments.

mitch is an R package for multi-contrast enrichment analysis. At it’s heart, it uses a rank-MANOVA based statistical approach to detect sets of genes that exhibit enrichment in the multidimensional space as compared to the background. The rank-MANOVA concept dates to work by Cox and Mann (https://doi.org/10.1186/1471-2105-13-S16-S12). mitch is useful for pathway analysis of profiling studies with one, two or more contrasts, or in studies with multiple omics profiling, for example proteomic, transcriptomic, epigenomic analysis of the same samples. mitch is perfectly suited for pathway level differential analysis of scRNA-seq data. We have an established routine for pathway enrichment of Infinium Methylation Array data (see vignette). The main strengths of mitch are that it can import datasets easily from many upstream tools and has advanced plotting features to visualise these enrichments.

A batteries-included toolkit for the GPU-accelerated OpenMM molecular simulation engine.

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.

This is a MobileViT (Small) model fine-tuned on the Processed Diabetic Retinopathy dataset.