janusx 1.0.13

A Suite of Genotyping Tools for Genome-Wide Association Study and Genomic Selection

JanusX

简体中文 | English | Zea Eureka

JanusX is a high-performance toolkit for quantitative genetics. It combines Rust-accelerated kernels (PyO3) with Python CLI workflows for GWAS, genomic selection (GS), post-analysis visualization, and variant-processing pipelines.

Overview

• Unified CLI entry: jx
• Core GWAS models: LM, LMM, FarmCPU
• Core GS models: GBLUP, rrBLUP, BayesA, BayesB, BayesCpi
• Streaming genotype reader for VCF/PLINK/TXT with low-memory workflows
• Integrated post-analysis tools: postgwas, postbsa, postgarfield
• Additional utilities: grm, pca, gmerge, fastq2vcf, sim, simulation

Installation

Requirements:

• Python >=3.9
• Rust toolchain (only needed when local wheel is unavailable and build from source is required)

PyPI (recommended)

pip install -U janusx
jx -h

Latest GitHub version

pip install --upgrade --force-reinstall --no-cache-dir git+https://github.com/FJingxian/JanusX.git
jx -h

Docker (source image build)

git clone https://github.com/FJingxian/JanusX.git
cd JanusX
docker build -t janusx:latest .
docker run --rm janusx:latest -h

In-place update for an installed JanusX

jx --update

CLI Modules

Genome-wide Association Studies (GWAS)

• grm: build genomic relationship matrix from genotype (-vcf or -bfile)
• pca: PCA for population structure from genotype, GRM prefix, or existing PCA prefix
• gwas: run genome-wide association analysis (LM, LMM, fastLMM, FarmCPU)
• postgwas: post-process GWAS results (Manhattan/QQ/annotation/merge/LD views)

Genomic Selection (GS)

• gs: genomic prediction and model-based selection

GARFIELD

• garfield: random-forest based marker-trait association
• postgarfield: summarize and visualize GARFIELD outputs

Bulk Segregation Analysis (BSA)

• postbsa: post-process and visualize BSA results

Pipeline and Utility

• fastq2vcf: variant-calling pipeline from FASTQ to VCF
• gmerge: merge genotype/variant datasets

Benchmark

• sim: quick simulation workflow
• simulation: extended simulation and benchmarking workflow

For full options, run:

jx <module> -h

Quick Start

1) GWAS

jx gwas -vcf example/mouse_hs1940.vcf.gz -p example/mouse_hs1940.pheno -lmm -plot -o test

2) Post-GWAS (single result)

jx postgwas -gwasfile test/mouse_hs1940.test0.lmm.tsv -manh -qq -threshold 1e-6 -o testpost

3) Post-GWAS merge plot (multiple GWAS files in one figure)

jx postgwas \
  -gwasfile testb/mouse_hs1940.test0.add.lmm.tsv \
  -merge testb/mouse_hs1940.test0.dom.lmm.tsv \
  -merge testb/mouse_hs1940.test0.het.lmm.tsv \
  -merge testb/mouse_hs1940.test0.rec.lmm.tsv \
  -manh -qq -o testpost

4) Genomic Selection

jx gs -vcf example/mouse_hs1940.vcf.gz -p example/mouse_hs1940.pheno -GBLUP -cv 5 -plot -o testgs

5) Post-BSA

jx postbsa -file QBproject/5.table/all.tsv -b1 JS -b2 JR -window 1 -format pdf -o QBproject/5.table

6) FASTQ to VCF pipeline

jx fastq2vcf -r /path/ref.fa -i /path/fastq_dir -w /path/workdir -b nohup -m 2

Input File Formats

Phenotype file

Tab-delimited. First column is sample ID, remaining columns are traits.

sample	trait1	trait2
id1	10.5	0.85
id2	12.3	1.00
id3	8.6	0.92

Genotype files

• VCF: .vcf or .vcf.gz (-vcf)
• PLINK: .bed/.bim/.fam prefix (-bfile)
• Text genotype matrix (-file, selected modules):
  • header row is sample IDs
  • subsequent rows are SNP-major numeric genotype values

Example (text matrix):

id1	id2	id3	id4
0	1	2	0
2	2	1	0
0	0	1	1

GWAS optional matrices

• GRM (-k/--grm in gwas): numeric matrix aligned to genotype sample order
• Q/PC covariate (-q/--qcov in gwas): PC count or covariate matrix file
• Covariate (-c/--cov in gwas): numeric matrix aligned to genotype sample order

All matrix files should be numeric-only, space/tab-delimited, without row/column headers.

GRM/PCA file conventions (updated)

• jx grm accepts genotype input only: -vcf or -bfile
• jx grm outputs:
  • default text mode: {prefix}.grm.txt and {prefix}.grm.txt.id
  • with --npy: {prefix}.grm.npy and {prefix}.grm.npy.id
• jx pca inputs (mutually exclusive):
  • genotype: -vcf or -bfile (compute PCA from genotype)
  • GRM prefix: -k/--grm <prefix> (expects {prefix}.grm.id and {prefix}.grm.txt or {prefix}.grm.npy)
  • existing PCA prefix: -q/--qcov <prefix> (expects {prefix}.eigenval and {prefix}.eigenvec, visualization-only mode)

Common Output Naming

• gwas: {prefix}.{trait}.{model}.tsv and optional {prefix}.{trait}.{genetic_model}.{model}.svg
• postgwas: {prefix}.manh.{format}, {prefix}.qq.{format}, {prefix}.{thr}.anno.tsv
• postgwas merge mode: {prefix}.merge.manh.{format}, {prefix}.merge.qq.{format}
• postbsa:
  • {prefix}.{bulk1}vs{bulk2}.snpidx.{format}
  • {prefix}.{bulk1}vs{bulk2}.bsa.{format}
  • {prefix}.{bulk1}vs{bulk2}.smooth.tsv
  • {prefix}.{bulk1}vs{bulk2}.thr.tsv
• gs: {prefix}.{trait}.gs.tsv

Python API (selected)

from janusx.gfreader import load_genotype_chunks

for geno, sites in load_genotype_chunks("example/mouse_hs1940.vcf.gz", chunk_size=50000):
    # geno: (m, n) float32 genotype chunk
    # sites: variant metadata list
    pass

from janusx.gtools.wgcna import cor, adj, tom, cluster

Citation

@article {FuJanusX,
  title = {JanusX: an integrated and high-performance platform for scalable genome-wide association studies and genomic selection},
  author = {Fu, Jingxian and Jia, Anqiang and Wang, Haiyang and Liu, Hai-Jun},
  year = {2026},
  doi = {10.64898/2026.01.20.700366},
  publisher = {Cold Spring Harbor Laboratory},
  URL = {https://www.biorxiv.org/content/early/2026/01/23/2026.01.20.700366},
  journal = {bioRxiv}
}

License

MIT License. See LICENSE.