Quickstart

The top-level package exposes the most common readers and data containers.

import snputils as su

snpobj = su.read_snp("your.vcf.gz")

read_snp dispatches from the file extension and returns a snputils.SNPObject.

bed = su.read_bed("cohort.bed")
pgen = su.read_pgen("cohort.pgen")
vcf = su.read_vcf("cohort.vcf.gz")

Local and Global Ancestry

lai = su.read_msp("local_ancestry.msp")
adm = su.read_admixture("global_ancestry")

Local ancestry files are represented as snputils.LocalAncestryObject; global ancestry and ADMIXTURE-style outputs use snputils.GlobalAncestryObject.

Phenotypes and IBD

phen = su.PhenotypeReader("phenotypes.tsv").read()
ibd = su.read_ibd("hap.ibd")

Phenotype data are represented by snputils.PhenotypeObject or snputils.MultiPhenotypeObject. Identity-by-descent segments are represented by snputils.IBDObject.

Analysis and Visualization

from snputils.processing import PCA
from snputils.stats import allele_freq_stream
from snputils.visualization import scatter

pca = PCA(n_components=2)
coords = pca.fit_transform(snpobj)

freq = allele_freq_stream("cohort.pgen", chunk_size=50_000)
fig = scatter(coords[:, 0], coords[:, 1])

For complete workflows, see the tutorials.