negative-binomial

Companion repository to Lause, Berens & Kobak (2021): "Analytic Pearson residuals for normalization of single-cell RNA-seq UMI data", Genome Biology

Probabilistic outlier identification for bulk RNA sequencing data

NewWave

Various Fortran codes

Code for fitting a negative binomial distribution in Python

Negative binomial distributed pseudorandom numbers.

Geographically Weighted Negative Binomial Regression in Python - local spatial modeling for over dispersed count data.

Research code and reproducibility materials for a mobility-informed SIR model using subway ridership data, particle smoothing, and mobility-reduction scenarios to evaluate influenza transmission and the instantaneous reproduction number.

High-precision ribosome pause detection tool utilizing Negative Binomial modeling to optimize Z-scores and extract ML-ready contextual features from Ribo-seq data.

The DOTNB repository is a collection of code files that implement DOTNB across several programming languages. The DOTNB is the distribution for the Difference Of Two Negative Binomial distributions, i.e., Z=X-Y ~ DOTNB (λ_1,λ_2,p_1,p_2), where X ~ NB(λ_1,p_1 ) and Y ~ NB(λ_2,p_2 ).

Create an iterator for generating pseudorandom numbers drawn from a negative binomial distribution.

纯 Python 标准库从零实现 RNA-seq 差异表达分析（归一化/离散度收缩/负二项 GLM/Wald/BH），与官方 DESeq2 对拍：log2FC r=0.9997、显著集 Jaccard=0.90。

Football prop forecasting engine. Post-competition review found the calibration layer was making forecasts worse, removing it is documented, measured, and enforced by tests.

Create an array containing pseudorandom numbers drawn from a negative binomial distribution.

RNS-Seq Count Model Explorer

The DEGage2 package works to identify differentially expressed genes (DEGs) on bulk RNA-seq data through utilization of DOTNB

DEGage is a novel model-based method for gene differential expression analysis between two groups of scRNA-seq count data. It employs a novel family of discrete distributions for describing the difference of two NB distributions (named DOTNB).

This project analyzes species observations from the National Park Service’s biodiversity database to examine how species category, nativeness, and abundance classification influence richness across U.S. National Parks.

Final project in Safety Management: analytics and predictive modeling for occupational incidents. Includes EDA, logistic regression, Poisson/Negative Binomial with overdispersion checks, ROC/AUC, and prediction exercises.

過分散と地点クラスタを扱う階層計数モデルの合成データ検証 — Synthetic validation of hierarchical count models for overdispersion and site clustering