Portfolio

Dai

Personal projects, packages, and tools.

Packages and tools

Current public repositories with documentation where available

R package

NestedWGCNA

R package for two-stage gene co-expression analysis.

Coarse-grained module discovery followed by GenFocus-normalized fine-grained module discovery in a nested workflow.

R package

uccdf

R package for typed consensus clustering.

Mixed-type data-frame clustering with schema inference, resampling, null testing, supported K selection, and sample-level assignments.

R package

scdown

R package for annotated single-cell RNA-seq downstream analysis.

A narrow package for cluster maps, marker discovery, annotation checks, signature summaries, and ligand-receptor ranking.

R package

cce

R package for counterfactual comparator analysis.

Standardized workflows for survival comparisons, SOC scenario planning, diagnostics, and export contracts in oncology-style analyses.

Python package

litmap

Python package for reproducible literature mapping.

A package and documentation scaffold for explicit run artifacts, cluster mapping, and review-ready literature analysis outputs.

Python package

trait2gene

Python package and CLI for trait-to-gene prioritization.

A PoPS-oriented workflow that standardizes resource resolution, MAGMA handoff, feature preparation, locus prioritization, and reporting.

Python package

turbocellatlas

Python package for atlas-scale single-cell retrieval.

Compressed candidate generation with exact reranking, benchmark artifacts, and documentation for computational and wet-lab users.

Python package

histgraph

Python package for histology-to-spatial-biology modeling.

A pipeline-oriented package for whole-slide histology workflows, graph construction, teacher models, and breast-cancer modeling on compute servers.

Python tool

oncoscape

HPC-oriented spatial pathology scaffold.

A script-first contract for breast cancer spatial workflows, upstream map generation, and downstream biomarker-model handoff.

R package

statsguider

R package for guided statistical test selection.

A narrow workflow that recommends and runs appropriate simple tests from study-design choices in a data frame.

R package

expranno

R package for RNA-seq workflows.

Expression-matrix based workflows for annotation, metadata integration, deconvolution, and signature analysis.

R package

heteff

R package for heterogeneous effect estimation.

Generalized random forest workflows for observational, survival, and instrumental-variable settings.