Source code for kdmsnakemake

from __future__ import division, print_function, absolute_import
from collections import defaultdict
import re

__version__ = "0.1.0"
__all__ = ["make_regions", "make_chromosomes"]


def parsefai(fai):
    """Parses a fasta index, yielding (chomosome, length) pairs"""
    with open(fai) as fh:
        for l in fh:
            cname, clen, _, _, _ = l.split()
            clen = int(clen)
            yield cname, clen


[docs]def make_regions(rdict, window=1e6): """Splits a reference into `window` sized windows. Makes a list of regions for each reference in a dict of {refname: refpath: entries. :param rdict: dict of {refname: refpath, ...} :param window: Size of windows :returns: dict of {refname: {region: coordinates, ...}, ...} """ window = int(window) ret = {} for refname, refpath in rdict.items(): fai = refpath+".fai" windows = [] curwin = [] curwinlen = 0 for cname, clen in parsefai(fai): if clen < window: curwinlen += clen reg = "{}:1-{}".format(cname, clen) curwin.append(reg) if curwinlen > window: windows.append(curwin) curwin = [] curwinlen = 0 else: for start in range(0, clen, window): wlen = min(clen - start, window) windows.append(["{}:{}-{}".format(cname, start, start+wlen)]) if len(curwin) > 0: windows.append(curwin) ref = dict() for i, w in enumerate(windows): wname = "W{:05d}".format(i) ref[wname] = w ret[refname] = ref return ret
[docs]def make_chromosomes(rdict, chrom_regex="^chr"): """Splits a reference into chromosome chunks Makes a list of regions for each reference in a dict of {refname: refpath} entries. Sequences matching `chrom_regex` appear as their own entires, all other sequences appear under a "scaffols" pseudo-chromosome. :param rdict: dict of {refname: refpath, ...} :param chrom_regex: Regular expression that (case-insensitively) matches proper chromosomes. :returns: dict of {refname: {chromosome_set: [refseq_name, ...]}, ...} """ ret = {} ischrom = re.compile(chrom_regex, re.IGNORECASE) for refname, refpath in rdict.items(): fai = refpath + ".fai" ref = dict() scafs = [] for cname, clen in parsefai(fai): if ischrom.match(cname) is not None: ref[cname] = [cname] else: scafs.append(cname) ref["scaffolds"] = scafs ret[refname] = ref return ret