Find open-source science resources

Transform arXiv papers into Beamer slides using LLMs

This tutorial involves the use of a multilayer AutoEncoder (AE) for feature extraction and pattern recognition by analyzing Molecular Dynamic Simulations, step by step, using the BioExcel Building Blocks library (biobb)

An elaborate molecular evolutionary framework that facilitates straightforward simulation of codon genetic sequences subjected to different degrees and/or patterns of Darwinian selection. The model is built upon the fitness landscape paradigm of Sewall Wright, as popularised by the mutation-selection model of Halpern and Bruno. This enables realistic evolutionary process of living organisms to be reproducible seamlessly. For example, an Ornstein-Uhlenbeck fitness update algorithm is incorporated herein. Consequently, otherwise complex biological processes, such as the effect of the interplay between genetic drift and fitness landscape fluctuations on the inference of diversifying selection, may now be investigated with minimal effort. Frequency-dependent and stochastic fitness landscape update techniques are available.

The tidySummarizedExperiment package provides a set of tools for creating and manipulating tidy data representations of SummarizedExperiment objects. SummarizedExperiment is a widely used data structure in bioinformatics for storing high-throughput genomic data, such as gene expression or DNA sequencing data. The tidySummarizedExperiment package introduces a tidy framework for working with SummarizedExperiment objects. It allows users to convert their data into a tidy format, where each observation is a row and each variable is a column. This tidy representation simplifies data manipulation, integration with other tidyverse packages, and enables seamless integration with the broader ecosystem of tidy tools for data analysis.

CCPlotR is an R package for visualising results from tools that predict cell-cell interactions from single-cell RNA-seq data. These plots are generic and can be used to visualise results from multiple tools such as Liana, CellPhoneDB, NATMI etc.

Provides customized print methods for 'SummarizedExperiment' objects to enhance readability and usability within a tidy workflow. It offers consistent, tidyverse-aligned console displays, including alternative tibble abstractions for large genomic data to improve discoverability and interpretation. The package also includes unified, contextual messaging utilities intended for the 'tidyomics' ecosystem.

Universal scientific research intelligence covering 50+ disciplines, repositioning LLMs as cross-disciplinary generators with human experts as verifiers; 30B model outperforms Claude Opus and GPT on 5 research benchmarks

RolDE detects longitudinal differential expression between two conditions in noisy high-troughput data. Suitable even for data with a moderate amount of missing values.RolDE is a composite method, consisting of three independent modules with different approaches to detecting longitudinal differential expression. The combination of these diverse modules allows RolDE to robustly detect varying differences in longitudinal trends and expression levels in diverse data types and experimental settings.

ChemmineOB provides an R interface to a subset of cheminformatics functionalities implemented by the OpelBabel C++ project. OpenBabel is an open source cheminformatics toolbox that includes utilities for structure format interconversions, descriptor calculations, compound similarity searching and more. ChemineOB aims to make a subset of these utilities available from within R. For non-developers, ChemineOB is primarily intended to be used from ChemmineR as an add-on package rather than used directly.

ChemmineR is a cheminformatics package for analyzing drug-like small molecule data in R. Its latest version contains functions for efficient processing of large numbers of molecules, physicochemical/structural property predictions, structural similarity searching, classification and clustering of compound libraries with a wide spectrum of algorithms. In addition, it offers visualization functions for compound clustering results and chemical structures.

EpiDISH is a R package to infer the proportions of a priori known cell-types present in a sample representing a mixture of such cell-types. Right now, the package can be used on DNAm data of blood-tissue of any age, from birth to old-age, generic epithelial tissue and breast tissue. Besides, the package provides a function that allows the identification of differentially methylated cell-types and their directionality of change in Epigenome-Wide Association Studies.

msPurity R package was developed to: 1) Assess the spectral quality of fragmentation spectra by evaluating the "precursor ion purity". 2) Process fragmentation spectra. 3) Perform spectral matching. What is precursor ion purity? -What we call "Precursor ion purity" is a measure of the contribution of a selected precursor peak in an isolation window used for fragmentation. The simple calculation involves dividing the intensity of the selected precursor peak by the total intensity of the isolation window. When assessing MS/MS spectra this calculation is done before and after the MS/MS scan of interest and the purity is interpolated at the recorded time of the MS/MS acquisition. Additionally, isotopic peaks can be removed, low abundance peaks are removed that are thought to have limited contribution to the resulting MS/MS spectra and the isolation efficiency of the mass spectrometer can be used to normalise the intensities used for the calculation.

High-throughput omics data are often affected by systematic biases introduced throughout all the steps of a clinical study, from sample collection to quantification. Normalization methods aim to adjust for these biases to make the actual biological signal more prominent. However, selecting an appropriate normalization method is challenging due to the wide range of available approaches. Therefore, a comparative evaluation of unnormalized and normalized data is essential in identifying an appropriate normalization strategy for a specific data set. This R package provides different functions for preprocessing, normalizing, and evaluating different normalization approaches. Furthermore, normalization methods can be evaluated on downstream steps, such as differential expression analysis and statistical enrichment analysis. Spike-in data sets with known ground truth and real-world data sets of biological experiments acquired by either tandem mass tag (TMT) or label-free quantification (LFQ) can be analyzed.

Open-source AI research assistant for biomedicine — chat to run RNA-seq, drug discovery, clinical analysis, and more; built on OpenClaw and Claude Code with 140 K-Dense scientific skills, real-time dashboard, and RStudio/JupyterLab integration (xjtulyc, 669+ stars, 2026)

ExpoRiskR provides tools for exposure-aware multi-omics risk modeling in translational and environmental health studies. The package aligns sample identifiers across exposure and multi-omics blocks, performs lightweight preprocessing, and fits exposure-adjusted association models to build interpretable microbe–metabolite networks. It also computes simple exposure perturbation summaries and generates publication-ready visualizations. Workflows support both matrix-based inputs and SummarizedExperiment objects.

Provide utilities for ROC, with microarray focus.

Provides tools for simulating copy-number alteration (CNA) profiles, applying a non-decimated Haar wavelet transform to genomic signals, and extracting wavelet-derived features for use in supervised learning. Multiple machine learning methods including lasso and elastic-net regularisation, random forest, partial least squares, neural networks and k-nearest neighbours are implemented to train predictive models from genomic feature vectors. The workflow enables end-to-end analysis from CNA simulation to feature extraction and classification.

maeparser is a parser for Schrodinger Maestro files.

fourSynergy is an ensemble algorithm leveraging synergies among the existing 4C-seq algorithms r3C-seq, peakC, r.4cker and fourSig. It uses a weighted voting approach to perform improved interaction calling. fourSynergy supports also differential interaction calling.

Genetic variant annotation and effect prediction toolbox.

Visualization of next generation sequencing (NGS) data is essential for interpreting high-throughput genomics experiment results. 'GenomicPlot' facilitates plotting of NGS data in various formats (bam, bed, wig and bigwig); both coverage and enrichment over input can be computed and displayed with respect to genomic features (such as UTR, CDS, enhancer), and user defined genomic loci or regions. Statistical tests on signal intensity within user defined regions of interest can be performed and represented as boxplots or bar graphs. Parallel processing is used to speed up computation on multicore platforms. In addition to genomic plots which is suitable for displaying of coverage of genomic DNA (such as ChIPseq data), metagenomic (without introns) plots can also be made for RNAseq or CLIPseq data as well.

This tutorial aims to illustrate the process of analyzing a membrane molecular dynamics (MD) simulation, step by step, using the BioExcel Building Blocks (biobb)

This tutorial aims to illustrate the process of protein-protein docking, step by step, using HADDOCK3 and the BioExcel Building Blocks (biobb)

This tutorial aims to illustrate the process of checking a molecular structure before using it as an input for a Molecular Dynamics simulation, step by step, using the BioExcel Building Blocks (biobb).

This BioExcel Building Blocks library (BioBB) workflow provides a pipeline to setup DNA structures for the Ascona B-DNA Consortium (ABC) members. It follows the work started with the NAFlex tool to offer a single, reproducible pipeline for structure preparation, ensuring reproducibility and coherence between all the members of the consortium.

This tutorial aims to illustrate the process of generating protein conformational ensembles from 3D structures using Coarse-Grained tools from the FlexServ server and analysing its molecular flexibility

This tutorial aims to illustrate the process of setting up a simulation system containing a protein in complex with a ligand, step by step, using the BioExcel Building Blocks library (biobb). The particular example used is the T4 lysozyme L99A/M102Q protein (PDB code 3HTB), in complex with the 2-propylphenol small molecule (3-letter Code JZ4).

This tutorial aims to illustrate the process of setting up a simulation system containing a protein, step by step, using the BioExcel Building Blocks library (biobb). The particular example used is the Lysozyme protein (PDB code 1AKI).

This tutorial aims to illustrate the process of computing a conformational transition between two known structural conformations of a protein, step by step, using the BioExcel Building Blocks (biobb).

This tutorial aims to illustrate the process of protein-ligand docking, step by step, using the BioExcel Building Blocks library (biobb).

This tutorial aims to illustrate the process of ligand parameterization for a small molecule, step by step, using the BioExcel Building Blocks library (biobb). The particular example used is the Ibuprofen small compound (3-letter code IBP, Drugbank code DB01050), a non-steroidal anti-inflammatory drug (NSAID) derived from propionic acid.

Structure-aware protein language model using 3D structural vocabulary (Foldseek) for joint sequence-structure pretraining, achieving SOTA on protein engineering and fitness prediction benchmarks (ICML 2024, Westlake University & Repl)

This package addresses the mean-variance relationship in spatially resolved transcriptomics data. Precision weights are generated for individual observations using Empirical Bayes techniques. These weights are used to rescale the data and covariates, which are then used as input in spatially variable gene detection tools.

Scientific foundation model and AI research copilot for idea generation, cross-disciplinary connection discovery, and hypothesis formation; trained with a decoupled reward-comment RL architecture and achieves GPT-4o-competitive novelty/rationale on STEM and social-science idea-generation benchmarks (270+ stars, MIT License, 2026)

This tutorial aims to illustrate the process of computing classical molecular interaction potentials from protein structures step by step, using the BioExcel Building Blocks library (biobb)

This tutorial aims to illustrate how to compute a fast-growth mutation free energy calculation, step by step, using the BioExcel Building Blocks library (biobb). The particular example used is the Staphylococcal nuclease protein (PDB code 1STN), a small, minimal protein, appropriate for a short tutorial.

Bilingual protein language model translating between protein sequence and structure, finetuned from ProtT5-XL on 17M AlphaFoldDB structures using Foldseek's 3Di structural alphabet, enabling sequence-to-structure prediction, structure-to-sequence inverse folding, and unified protein representation learning (RostLab, 310+ stars)

systemPipeR is a workflow management environment for reproducible data analysis that integrates R with command-line software. It enables researchers to design, execute, and report complex workflows on local machines and HPC systems. The framework combines R-based analysis with external tools through a Common Workflow Language (CWL) interface, manages workflow dependencies and restart capabilities, and automatically generates reproducible scientific analysis reports. The companion package systemPipeRdata provides ready-to-use workflow templates that simplify workflow setup and customization. Alternatively, workflow templates can be loaded from dedicated GitHub repositories.

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)

cogeqc aims to facilitate systematic quality checks on standard comparative genomics analyses to help researchers detect issues and select the most suitable parameters for each data set. cogeqc can be used to asses: i. genome assembly and annotation quality with BUSCOs and comparisons of statistics with publicly available genomes on the NCBI; ii. orthogroup inference using a protein domain-based approach and; iii. synteny detection using synteny network properties. There are also data visualization functions to explore QC summary statistics.

First benchmark for automatic video generation from scientific papers (NeurIPS 2025)

RFantibody is a pipeline for structure-based de novo antibody and nanobody design, integrating backbone design with RFdiffusion, sequence design with ProteinMPNN, and structure prediction with RoseTTAFold2. It provides a comprehensive toolset for generating and filtering high-quality antibody designs.

Interactive personal genome analysis toolkit using Claude Code and Python. Parses raw genotyping data from consumer DNA services and analyzes SNPs across 17 categories including health risks, pharmacogenomics, ancestry, and nutrition, with a terminal-style HTML dashboard.

Chemoinformatics and drug discovery section in deeplearning-biology repo.

Open-source platform for building, extending, and experimenting with scientific agents, providing modular agent construction tools and standardized evaluation pipelines for accelerating autonomous scientific discovery research (748+ stars, MIT License)

The GENESIS package provides methodology for estimating, inferring, and accounting for population and pedigree structure in genetic analyses. The current implementation provides functions to perform PC-AiR (Conomos et al., 2015, Gen Epi) and PC-Relate (Conomos et al., 2016, AJHG). PC-AiR performs a Principal Components Analysis on genome-wide SNP data for the detection of population structure in a sample that may contain known or cryptic relatedness. Unlike standard PCA, PC-AiR accounts for relatedness in the sample to provide accurate ancestry inference that is not confounded by family structure. PC-Relate uses ancestry representative principal components to adjust for population structure/ancestry and accurately estimate measures of recent genetic relatedness such as kinship coefficients, IBD sharing probabilities, and inbreeding coefficients. Additionally, functions are provided to perform efficient variance component estimation and mixed model association testing for both quantitative and binary phenotypes.

Our scLANE model uses truncated power basis spline models to build flexible, interpretable models of single cell gene expression over pseudotime or latent time. The modeling architectures currently supported are Negative-binomial GLMs, GEEs, & GLMMs. Downstream analysis functionalities include model comparison, dynamic gene clustering, smoothed counts generation, gene set enrichment testing, & visualization.

Mass spectrometry (MS) data backend supporting import and export of MS/MS spectra data from Mascot Generic Format (mgf) files. Objects defined in this package are supposed to be used with the Spectra Bioconductor package. This package thus adds mgf file support to the Spectra package.

Mass spectrometry (MS) data backend supporting import and handling of MS/MS spectra from NIST MSP Format (msp) files. Import of data from files with different MSP *flavours* is supported. Objects from this package add support for MSP files to Bioconductor's Spectra package. This package is thus not supposed to be used without the Spectra package that provides a complete infrastructure for MS data handling.

First benchmark evaluating LLMs' ability to rediscover scientific laws through interactive experimentation across 324 tasks in 12 physics domains, featuring memorization-resistant metaphysical shifts of canonical laws (HKUST)