The sexassign.py script uses counts of reads mapping to the X chromosome
and autosomes to determine the genetic sex of an individual. Under the
assumption that the number of reads obtained should reflect the chromosome
copy numbers and chromosome lengths, two binomial models are constructed,
one for males and one for females. A likelihood ratio test is then used
to distinguish between the two models. This is effective with as few as
5000 mapped shotgun reads.
The sexassign.py script requires python, numpy, scipy, and matplotlib.
Read counts must first be extracted from mapped BAM files using
samtools idxstats, and sexassign.py can then be used output a
tab-separated table of sex assignments, and to plot read dosages.
$ for i in *.bam; do samtools idxstats ${i} > ${i}.idxstats; done
$ sexassign.py -w -o sexes.pdf *.idxstats > sexes.txt
Further information can be obtained from the help output.
$ sexassign.py -h
usage: sexassign.py [-h] [-w] [-o OPDF] [-n MIN_READS] [-l MIN_LENGTH]
[-X CHRX_LIST] [-e EXCLUDE_LIST] [-i INCLUDE_LIST] [-p]
infiles [infiles ...]
sex determination from read dosages
positional arguments:
infiles input file(s)
optional arguments:
-h, --help show this help message and exit
-w, --wide plot widescreen ratio (16x9) [False]
-o OPDF, --opdf OPDF output filename [out.pdf]
-n MIN_READS, --min-reads MIN_READS
exclude samples with fewer reads than this [5000]
-l MIN_LENGTH, --min-length MIN_LENGTH
exclude contigs shorter than this [1000000]
-X CHRX_LIST, --chrX-list CHRX_LIST
file containing list of X linked contigs, if not
specified, defaults to X, chrX, ChrX
-e EXCLUDE_LIST, --exclude-list EXCLUDE_LIST
file containing list of contigs to exclude, if not
specified, defaults to Y, chrY, ChrY, M, MT, Mt
-i INCLUDE_LIST, --include-list INCLUDE_LIST
file containing list of contigs to include, if not
specified, defaults to all contigs
-p, --pca Plot PCA of read dosages
If you find this script useful, please cite: Gower et al. (2019), https://doi.org/10.1073/pnas.1903275116