view autogenJB2.py @ 36:5f39f745682f draft

planemo upload for repository https://github.com/usegalaxy-eu/temporary-tools/tree/master/jbrowse2 commit 80b849766a962bac4bd0bb8cb69c118cc42699cd-dirty
author fubar
date Thu, 29 Feb 2024 00:47:58 +0000
parents 15da358c3108
children 07849bf248e3
line wrap: on
line source

import argparse
import logging
import sys

from jbrowse2 import JbrowseConnector as jbC

logging.basicConfig(level=logging.debug)
log = logging.getLogger("jbrowse")


def makeDefaultLocation():

    refName = jc.genome_firstcontig
    defloc  = "%s:100..10000" % refName
    print ('defloc',defloc)
    return defloc


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="", epilog="")
    parser.add_argument("--sessName", help="Session name", default="AutoJBrowse")
    parser.add_argument(
        "--trackmeta",
        help="Repeatable of 'filename, filext,filepath, ... ,' for JBrowse2 tracks",
        default=[],
        action="append",
    )
    parser.add_argument(
        "--referencemeta",
        help="Repeatable 'filename, filext, filepath, ... ,' for JBrowse2 reference tracks",
        default=[],
        action="append",
    )
    parser.add_argument(
        "--pafmeta",
        help="Repeatable. Each is a 'filename, filext, filepath, ... ,' for a JBrowse2 paf track",
        default=[],
        action="append",
    )
    parser.add_argument(
        "--pafreferencemeta",
        help="Repeatable. Each is a 'pafname,filepath,refname,filepath,refname....'. Every paf must have a corresponding one ",
        default=[],
        action="append",
    )

    parser.add_argument("--version", "-V", action="version", version="%(prog)s 2.10.2")
    parser.add_argument("--outdir", help="Output directory", required=True)
    args = parser.parse_args()
    sessName = args.sessName
    # --trackmeta $jbrowseme[$key],$jbrowseme[$key].ext,'$key'
    trackList = [x.strip().split(",") for x in args.trackmeta if x > '']
    refList = [x.strip().split(",") for x in args.referencemeta if x > '']
    print("tracklist = %s\nreflist = %s" % (trackList,refList))
    if len(refList) > 0:
        listgenomes = [f for f in refList if f[1] in ['fasta', 'fasta.gz']]
        # assume no pafs here
        print('genomes=%s' % listgenomes)
        if len(listgenomes) > 0:
            genome_paths = [
                x[0] for x in listgenomes
            ]  # expect genome_1_genomename.fasta etc
            genome_names = [x[2] for x in listgenomes]
            jc = jbC(
                outdir=args.outdir,
                genomes=[
                    {
                        "path": x,
                        "meta": {
                            "name": genome_names[i],
                            "dataset_dname": genome_names[i],
                        },
                    }
                    for i, x in enumerate(genome_paths)
                ],
            )
            sys.stdout.write('$$$ genome_paths:%s genome_names: %s' % (genome_paths,genome_names))
            jc.process_genomes()
            default_session_data = {
                "visibility": {
                    "default_on": [],
                    "default_off": [],
                },
                "style": {},
                "style_labels": {},
            }

            listtracks = trackList
            # foo.paf must have a foo_paf.fasta or fasta.gz to match
            tnames = [x[2] for x in listtracks]
            texts = [x[1] for x in listtracks]
            for i, track in enumerate(listtracks):
                tpath, trext, trackname = track[:3]
                if trext == "paf":
                    refname = trackname + "_paf.fasta"
                    refdat = [x[2] for x in listtracks if x[2] == refname]
                    if not refdat:
                        jc.logging.warn(
                            "!! No reference file %s corresponding to paf file %s found. Not building - there must be a corresponding fasta for each paf"
                            % (refname, trackname)
                        )
                        sys.exit(3)
                    else:
                        track_conf = {
                            "conf": {
                                "options": {
                                    "paf": {"genome": refdat, "genome_label": trackname}
                                }
                            }
                        }
                elif trext == 'bam':
                     track_conf = {"conf": {"options": {"bam": {"bam_indices": {"bam_index": track[3]}}}}}
                elif trext == 'cram':
                     track_conf = {"conf": {"options": {"cram": {"cram_indices": {"cram_index": track[3]}}}}}
                else:
                    track_conf = {}
                track_conf["format"] = trext
                track_conf["name"] = trackname
                track_conf["label"] = trackname
                track_conf["trackfiles"] = [(tpath, trext, trackname,{}),]
                track_conf["category"] = "Autogenerated"
                keys = jc.process_annotations(track_conf)

                if keys:
                    for key in keys:
                        if trext in ["bigwig", "gff3", "gff", "vcf", "maf",]:
                            default_session_data["visibility"]["default_on"].append(key)
                        else:
                            default_session_data["visibility"]["default_off"].append(key)
                        # if track_conf.get("style", None):
                        # default_session_data["style"][key] = track_conf[
                        # "style"
                        # ]  # TODO do we need this anymore?
                        # if track_conf.get("style_lables", None):
                        # default_session_data["style_labels"][key] = track_conf.get(
                        # "style_labels", None
                        # )
            # general_data = {
            # "analytics": root.find("metadata/general/analytics").text,
            # "primary_color": root.find("metadata/general/primary_color").text,
            # "secondary_color": root.find("metadata/general/secondary_color").text,
            # "tertiary_color": root.find("metadata/general/tertiary_color").text,
            # "quaternary_color": root.find("metadata/general/quaternary_color").text,
            # "font_size": root.find("metadata/general/font_size").text,
            # }
            # jc.add_general_configuration(general_data)
            trackconf = jc.config_json.get("tracks", None)
            if trackconf:
                jc.config_json["tracks"].update(jc.tracksToAdd)
            else:
                jc.config_json["tracks"] = jc.tracksToAdd
            jc.write_config()
            defLoc = makeDefaultLocation()
            default_session_data.update({"defaultLocation": defLoc, "session_name": sessName})
            track_conf.update(default_session_data)
            jc.add_default_session(default_session_data)
            # jc.text_index() not sure what broke here.
    else:
        print("!! empty collection supplied - nothing to process")