Find open-source science resources

A VCF Parser for Python.

This model may be overfit to some extent (see below). Try running this notebook on the datasets linked to in the notebook. See if you can figure out why the metrics differ so much on the datasets. Is it due to something like sequence similarity in the train/test split?

First vision-and-language foundation model for pathology AI, fine-tuned from CLIP on 249K image-caption pairs, enabling open-ended visual-semantic search and zero-shot diagnosis across histopathology (Pathology Foundation, 376+ stars)

Wrapper for RDKit's RunReactants to improve stereochemistry handling

+ Model Name: Med_English2Spanish + Model Type: Transformer-based Neural Machine Translation (NMT) Model + Task: English to Spanish Medical Translation

This model is a fine-tuned model based on the Llama 2_7b architecture. It has been specifically trained on a dataset comprising USMLE (United States Medical Licensing Examination) questions and answers, as well as conversations between doctors and patients.

MEM, Marker Enrichment Modeling, automatically generates and displays quantitative labels for cell populations that have been identified from single-cell data. The input for MEM is a dataset that has pre-clustered or pre-gated populations with cells in rows and features in columns. Labels convey a list of measured features and the features' levels of relative enrichment on each population. MEM can be applied to a wide variety of data types and can compare between MEM labels from flow cytometry, mass cytometry, single cell RNA-seq, and spectral flow cytometry using RMSD.

Git repo of useful single line commands.

Clonal cell groups share common mutations within cancer, precancer, and even clinically normal appearing tissues. The frequency and location of these mutations may predict prognosis and cancer risk. It has also been well established that certain genomic regions have increased sensitivity to acquiring mutations. Mutation-sensitive genomic regions may therefore serve as markers for predicting cancer risk. This package contains multiple functions to establish significantly mutated hotspots, compare hotspot mutation burden between samples, and perform exploratory data analysis of the correlation between hotspot mutation burden and personal risk factors for cancer, such as age, gender, and history of carcinogen exposure. This package allows users to identify robust genomic markers to help establish cancer risk.

This package takes a list of p-values resulting from the simultaneous testing of many hypotheses and estimates their q-values and local FDR values. The q-value of a test measures the proportion of false positives incurred (called the false discovery rate) when that particular test is called significant. The local FDR measures the posterior probability the null hypothesis is true given the test's p-value. Various plots are automatically generated, allowing one to make sensible significance cut-offs. Several mathematical results have recently been shown on the conservative accuracy of the estimated q-values from this software. The software can be applied to problems in genomics, brain imaging, astrophysics, and data mining.

Screen a bacterial assembly (contigs/CDS or proteins) for nucleotide or protein sequences. Pipeline that screens for presence of genes of interest (GOI) in bacterial assemblies. Generates multiple CSVs and plots that describe which genes are present and how variable their sequence is. Can use DNA or protein query sequences (GOIs) and DNA contigs/fastas or protein fastas as database (db) to search in.

A tool for the identification of differentially coexpressed links (DCLs) and differentially coexpressed genes (DCGs). DCLs are gene pairs with significantly different correlation coefficients under two conditions. DCGs are genes with significantly more DCLs than by chance.

Displaying sequence statistics for next-generation sequencing.

The epistack package main objective is the visualizations of stacks of genomic tracks (such as, but not restricted to, ChIP-seq, ATAC-seq, DNA methyation or genomic conservation data) centered at genomic regions of interest. epistack needs three different inputs: 1) a genomic score objects, such as ChIP-seq coverage or DNA methylation values, provided as a `GRanges` (easily obtained from `bigwig` or `bam` files). 2) a list of feature of interest, such as peaks or transcription start sites, provided as a `GRanges` (easily obtained from `gtf` or `bed` files). 3) a score to sort the features, such as peak height or gene expression value.

The ERSSA package takes user supplied RNA-seq differential expression dataset and calculates the number of differentially expressed genes at varying biological replicate levels. This allows the user to determine, without relying on any a priori assumptions, whether sufficient differential detection has been acheived with their RNA-seq dataset.

Methods and functionality to analyse flow data that is beyond the basic infrastructure provided by the flowCore package.

Differential expression analysis of RNA-seq using the Poisson-Tweedie (PT) family of distributions. PT distributions are described by a mean, a dispersion and a shape parameter and include Poisson and NB distributions, among others, as particular cases. An important feature of this family is that, while the Negative Binomial (NB) distribution only allows a quadratic mean-variance relationship, the PT distributions generalizes this relationship to any orde.

An open, extensible Python framework for GPU-accelerated alchemical free energy calculations.

mtag is a Python-based command line tool for jointly analyzing multiple sets of GWAS summary statistics as described by Turley et. al. (2018). It can also be used as a tool to meta-analyze GWAS results.

Methods for microarray analysis that take basic data types such as matrices and lists of vectors. These methods can be used standalone, be utilized in other packages, or be wrapped up in higher-level classes.

RegioneReloaded is a package that allows simultaneous analysis of associations between genomic region sets, enabling clustering of data and the creation of ready-to-publish graphs. It takes over and expands on all the features of its predecessor regioneR. It also incorporates a strategy to improve p-value calculations and normalize z-scores coming from multiple analysis to allow for their direct comparison. RegioneReloaded builds upon regioneR by adding new plotting functions for obtaining publication-ready graphs.

项目地址：https://github.com/iioSnail/chinesemedicalner

This is a code repository for the SIB - Swiss Institute of Bioinformatics CALIPHO group neXtProt project, which is a comprehensive human-centric discovery platform, that offers a integration of and navigation through protein-related data. CALIPHO is an interdisciplinary team which aims to use a variety of methodologies to help uncover the function of uncharacterized human proteins.

This is a Japanese RoBERTa base model pre-trained on academic articles in medical sciences collected by Japan Science and Technology Agency (JST).

I present a demo showcasing retinal vessel segmentation using the U-Net model, which is a well-known and widely used model in medical image segmentation. The model was trained on the DRIVE dataset, and the training process was conducted on Google Colab.

datasets: - UMLS

lipidr an easy-to-use R package implementing a complete workflow for downstream analysis of targeted and untargeted lipidomics data. lipidomics results can be imported into lipidr as a numerical matrix or a Skyline export, allowing integration into current analysis frameworks. Data mining of lipidomics datasets is enabled through integration with Metabolomics Workbench API. lipidr allows data inspection, normalization, univariate and multivariate analysis, displaying informative visualizations. lipidr also implements a novel Lipid Set Enrichment Analysis (LSEA), harnessing molecular information such as lipid class, total chain length and unsaturation.

HiCcompare provides functions for joint normalization and difference detection in multiple Hi-C datasets. HiCcompare operates on processed Hi-C data in the form of chromosome-specific chromatin interaction matrices. It accepts three-column tab-separated text files storing chromatin interaction matrices in a sparse matrix format which are available from several sources. HiCcompare is designed to give the user the ability to perform a comparative analysis on the 3-Dimensional structure of the genomes of cells in different biological states.`HiCcompare` differs from other packages that attempt to compare Hi-C data in that it works on processed data in chromatin interaction matrix format instead of pre-processed sequencing data. In addition, `HiCcompare` provides a non-parametric method for the joint normalization and removal of biases between two Hi-C datasets for the purpose of comparative analysis. `HiCcompare` also provides a simple yet robust method for detecting differences between Hi-C datasets.

Educational resource on performing RNA-seq analysis in the cloud using Amazon AWS cloud services. Topics include preparing the data, preprocessing, differential expression, isoform discovery, data visualization, and interpretation.

dStruct identifies differentially reactive regions from RNA structurome profiling data. dStruct is compatible with a broad range of structurome profiling technologies, e.g., SHAPE-MaP, DMS-MaPseq, Structure-Seq, SHAPE-Seq, etc. See Choudhary et al., Genome Biology, 2019 for the underlying method.

In recent years, pre-trained language models (PLMs) achieve the best performance on a wide range of natural language processing (NLP) tasks. While the first models were trained on general domain data, specialized ones have emerged to more effectively treat specific domains.

In recent years, pre-trained language models (PLMs) achieve the best performance on a wide range of natural language processing (NLP) tasks. While the first models were trained on general domain data, specialized ones have emerged to more effectively treat specific domains.

In recent years, pre-trained language models (PLMs) achieve the best performance on a wide range of natural language processing (NLP) tasks. While the first models were trained on general domain data, specialized ones have emerged to more effectively treat specific domains.

In recent years, pre-trained language models (PLMs) achieve the best performance on a wide range of natural language processing (NLP) tasks. While the first models were trained on general domain data, specialized ones have emerged to more effectively treat specific domains.

Easily submitting PBS jobs with script template. Multiple input files supported.

A suite of algorithms for matching position weight matrices (PWM) against DNA sequences. It features advanced matrix matching algorithms implemented in C++ that can be used to scan hundreds of matrices against chromosome-sized sequences in few seconds. MOODS can also process high-order PWMs with dependencies between adjacent positions and sequence variants such as SNPs, insertions and deletions.

A Jupyter book demonstrating working with biochemical data using the scikit-bio library for tasks such as sequence alignment and calculating Hamming distances.

R interface to the MELTING 5 program (https://www.ebi.ac.uk/biomodels/tools/melting/) to compute melting temperatures of nucleic acid duplexes along with other thermodynamic parameters.

ShinyÉPICo is a graphical pipeline to analyze Illumina DNA methylation arrays (450k or EPIC). It allows to calculate differentially methylated positions and differentially methylated regions in a user-friendly interface. Moreover, it includes several options to export the results and obtain files to perform downstream analysis.

Provide functions to obtain instrumentation data on processes in a unix environment. Parse output of a collectl run. Vizualize aspects of system usage over time, with annotation.

A Library for Deep Learning in Biology and Chemistry.

A deep learning framework (based on Chainer) with applications in Biology and Chemistry.

This is a comprehensive package to perform Tensor decomposition based unsupervised feature extraction. It can perform unsupervised feature extraction. It uses tensor decomposition. It is applicable to gene expression, DNA methylation, and histone modification etc. It can perform multiomics analysis. It is also potentially applicable to single cell omics data sets.

scCB2 is an R package implementing CB2 for distinguishing real cells from empty droplets in droplet-based single cell RNA-seq experiments (especially for 10x Chromium). It is based on clustering similar barcodes and calculating Monte-Carlo p-value for each cluster to test against background distribution. This cluster-level test outperforms single-barcode-level tests in dealing with low count barcodes and homogeneous sequencing library, while keeping FDR well controlled.

Convert between different data formats used by differential gene expression analysis tools.

This is an advanced version of TDbasedUFE, which is a comprehensive package to perform Tensor decomposition based unsupervised feature extraction. In contrast to TDbasedUFE which can perform simple the feature selection and the multiomics analyses, this package can perform more complicated and advanced features, but they are not so popularly required. Only users who require more specific features can make use of its functionality.