comparison jbrowse2.py @ 60:81d535970196 draft

planemo upload for repository https://github.com/usegalaxy-eu/temporary-tools/tree/master/jbrowse2 commit 403a35e1245fa5e62f4be6116a725b9e4d9c353a

59:f807e219cec3

 #!/usr/bin/env python
# change to accumulating all configuration for config.json based on the default from the clone
import argparse
import binascii
import datetime
import json
import logging

JB2VER = "v2.10.3"
# version pinned for cloning

TODAY = datetime.datetime.now().strftime("%Y-%m-%d")
GALAXY_INFRASTRUCTURE_URL = None

# version pinned for cloning

mapped_chars = {
    ">": "__gt__",
    "<": "__lt__",
    "'": "__sq__",

            genome_name = genome_node["meta"]["dataset_dname"].strip()
            if len(genome_name.split()) > 1:
                genome_name = genome_name.split()[0]
                # spaces and cruft break scripts when substituted
            if genome_name not in self.genome_names:
                # ignore dupes - can have multiple pafs with same references?
                fapath = genome_node["path"]
                if not useuri:
                    fapath = os.path.realpath(fapath)
                assem = self.make_assembly(fapath, genome_name, useuri)
                assemblies.append(assem)
                self.genome_names.append(genome_name)
                if self.genome_name is None:
                    self.genome_name = (
                        genome_name  # first one for all tracks - other than paf
                    )
                    self.genome_sequence_adapter = assem["sequence"]["adapter"]
                    self.genome_firstcontig = None
                    if not useuri:
                        fl = open(fapath, "r").readline()

                            if len(fl.split()) > 1:
                                self.genome_firstcontig = fl.split()[0].strip()
                            else:
                                self.genome_firstcontig = fl
                    else:
                        fl = urllib.request.urlopen(fapath + ".fai").readline()
                        if fl:  # is first row of the text fai so the first contig name
                            self.genome_firstcontig = (
                                fl.decode("utf8").strip().split()[0]
                            )
        if self.config_json.get("assemblies", None):
            self.config_json["assemblies"] += assemblies
        else:
            self.config_json["assemblies"] = assemblies

            faiLocation:
            uri: 'https://s3.amazonaws.com/jbrowse.org/genomes/GRCh38/fasta/GRCh38.fa.gz.fai',
            gziLocation:
            uri: 'https://s3.amazonaws.com/jbrowse.org/genomes/GRCh38/fasta/GRCh38.fa.gz.gzi',
        Cool will not be likely to be a good fit - see discussion at https://github.com/GMOD/jbrowse-components/issues/2438
        """
        tId = trackData["label"]
        # can be served - if public.
        # dsId = trackData["metadata"]["dataset_id"]
        # url = "%s/api/datasets/%s/display?to_ext=hic " % (self.giURL, dsId)
        useuri = trackData["useuri"].lower() == "yes"
        if useuri:
            uri = data
        else:
            uri = trackData["hic_url"]
        categ = trackData["category"]
        trackDict = {
            "type": "HicTrack",
            "trackId": tId,
            "name": uri,
            "assemblyNames": [self.genome_name],
            "category": [
                categ,
            ],
            "adapter": {
                "type": "HicAdapter",
                "hicLocation": uri,
            },
            "displays": [
                {
                    "type": "LinearHicDisplay",
                    "displayId": "%s-LinearHicDisplay" % tId,
                },
            ],
        }
        style_json = self._prepare_track_style(trackDict)
        trackDict["style"] = style_json
        self.tracksToAdd.append(trackDict)
        self.trackIdlist.append(tId)

    def add_maf(self, data, trackData):
        """

        self.tracksToAdd.append(trackDict)
        self.trackIdlist.append(tId)
        os.unlink(gff3)

    def add_bigwig(self, data, trackData):
        """ "type": "LinearWiggleDisplay",
        "configuration": {},
        "selectedRendering": "",
        "resolution": 1,
        "posColor": "rgb(228, 26, 28)",
        "negColor": "rgb(255, 255, 51)",
        "constraints": {}
        """
        useuri = trackData["useuri"].lower() == "yes"
        if useuri:
            url = data
        else:
            url = "%s.bigwig" % trackData["label"]

            # Return non-human label for use in other fields
            yield outputTrackConfig["label"]

    def add_default_session(self, default_data):
        """
        Add some default session settings: set some assemblies/tracks on/off
        """
        tracks_data = []

        # TODO using the default session for now, but check out session specs in the future https://github.com/GMOD/jbrowse-components/issues/2708

        # We need to know the track type from the config.json generated just before
        track_types = {}
        with open(self.config_json_file, "r") as config_file:
            config_json = json.load(config_file)
        if self.config_json:
            config_json.update(self.config_json)

        for track_conf in self.tracksToAdd:
            track_types[track_conf["trackId"]] = track_conf["type"]
            tId = track_conf["trackId"]
            #if tId in data["visibility"]["default_on"]:
            style_data = default_data["style"].get(tId,  None)
            if not style_data:
                logging.warn("### No style data in default data for %s" % tId)
                style_data = {"type": "LinearBasicDisplay"}
            if "displays" in track_conf:
                disp = track_conf["displays"][0]["type"]
                style_data["type"] = disp
            if track_conf.get("style_labels", None):

                    "displays": [style_data],
                }
            )
        # The view for the assembly we're adding
        view_json = {"type": "LinearGenomeView", "tracks": tracks_data}

        refName = None
        drdict = {
            "reversed": False,
            "assemblyName": self.genome_name,
            "start": 1,

        config_json["configuration"].update(config_data)
        self.config_json.update(config_json)
        with open(config_path, "w") as config_file:
            json.dump(self.config_json, config_file, indent=2)

    def clone_jbrowse(self, realclone=True):
        """Clone a JBrowse directory into a destination directory. This also works in Biocontainer testing now"""
        dest = self.outdir
        if realclone:
            self.subprocess_check_call(
                ["jbrowse", "create", dest, "-f", "--tag", f"{JB2VER}"]
            )

            for x in root.findall("metadata/genomes/genome")
        ],
    )
    jc.process_genomes()

    # .add_default_view() replace from https://github.com/abretaud/tools-iuc/blob/jbrowse2/tools/jbrowse2/jbrowse2.py
    default_session_data = {
        "visibility": {
            "default_on": [],
            "default_off": [],
        },

        trackfiles = track.findall("files/trackFile")
        if trackfiles:
            for x in track.findall("files/trackFile"):
                track_conf["label"] = x.attrib["label"]
                track_conf["useuri"] = x.attrib["useuri"]
                if is_multi_bigwig:
                    multi_bigwig_paths.append(
                        (
                            x.attrib["label"],

                    {},  # No metadata for multiple bigwig
                )
            )
        track_conf["category"] = track.attrib["cat"]
        track_conf["format"] = track.attrib["format"]
        if track.find("options/style"):
            track_conf["style"] = {
                item.tag: parse_style_conf(item) for item in track.find("options/style")
            }
        else:
            track_conf["style"] = {}
        if track.find("options/style_labels"):
            track_conf["style_labels"] = {
                item.tag: parse_style_conf(item)
                for item in track.find("options/style_labels")
            }

        track_conf["conf"] = etree_to_dict(track.find("options"))
        track_conf["category"] = track.attrib["cat"]
        track_conf["format"] = track.attrib["format"]
        keys = jc.process_annotations(track_conf)

        if keys:
            for key in keys:
                default_session_data["visibility"][
                    track.attrib.get("visibility", "default_off")
                ].append(key)
                if track_conf.get("style", None):
                    default_session_data["style"][key] = track_conf["style"]
                if track_conf.get("style_lables", None):
                    default_session_data["style_labels"][key] = track_conf.get(
                        "style_labels", None
                    )
    default_session_data["defaultLocation"] = root.find(
        "metadata/general/defaultLocation"
    ).text
    default_session_data["session_name"] = root.find(
        "metadata/general/session_name"

60:81d535970196

 #!/usr/bin/env python

import argparse
import binascii
import datetime
import json
import logging

JB2VER = "v2.10.3"
# version pinned for cloning

TODAY = datetime.datetime.now().strftime("%Y-%m-%d")
GALAXY_INFRASTRUCTURE_URL = None

mapped_chars = {
    ">": "__gt__",
    "<": "__lt__",
    "'": "__sq__",

            genome_name = genome_node["meta"]["dataset_dname"].strip()
            if len(genome_name.split()) > 1:
                genome_name = genome_name.split()[0]
                # spaces and cruft break scripts when substituted
            if genome_name not in self.genome_names:
                # pafs with shared references
                fapath = genome_node["path"]
                if not useuri:
                    fapath = os.path.realpath(fapath)
                assem = self.make_assembly(fapath, genome_name, useuri)
                assemblies.append(assem)
                self.genome_names.append(genome_name)
                if self.genome_name is None:
                    self.genome_name = (
                        genome_name  # first one for all tracks
                    )
                    self.genome_sequence_adapter = assem["sequence"]["adapter"]
                    self.genome_firstcontig = None
                    if not useuri:
                        fl = open(fapath, "r").readline()

                            if len(fl.split()) > 1:
                                self.genome_firstcontig = fl.split()[0].strip()
                            else:
                                self.genome_firstcontig = fl
                    else:
                        try:
                            fl = urllib.request.urlopen(fapath + ".fai").readline()
                        except:
                            fl = None
                        if fl:  # is first row of the text fai so the first contig name
                            self.genome_firstcontig = (
                                fl.decode("utf8").strip().split()[0]
                            )
                        else:
                            self.genome_firstcontig = None
        if self.config_json.get("assemblies", None):
            self.config_json["assemblies"] += assemblies
        else:
            self.config_json["assemblies"] = assemblies

            faiLocation:
            uri: 'https://s3.amazonaws.com/jbrowse.org/genomes/GRCh38/fasta/GRCh38.fa.gz.fai',
            gziLocation:
            uri: 'https://s3.amazonaws.com/jbrowse.org/genomes/GRCh38/fasta/GRCh38.fa.gz.gzi',
        Cool will not be likely to be a good fit - see discussion at https://github.com/GMOD/jbrowse-components/issues/2438


        """
        tId = trackData["label"]
        # can be served - if public.
        # dsId = trackData["metadata"]["dataset_id"]
        # url = "%s/api/datasets/%s/display?to_ext=hic " % (self.giURL, dsId)
        useuri = trackData["useuri"].lower() == "yes"
        if useuri:
            uri = data
        else:
            uri = "%s.hic" % trackData["label"]
            # slashes in names cause path trouble
            dest = os.path.join(self.outdir, uri)
            cmd = ["cp", data, dest]
            self.subprocess_check_call(cmd)
        categ = trackData["category"]
        trackDict = {
            "type": "HicTrack",
            "trackId": tId,
            "name":  trackData["name"],
            "assemblyNames": [self.genome_name],
            "category": [
                categ,
            ],
            "adapter": {
                "type": "HicAdapter",
                "hicLocation": { "uri": uri }
            }
        }
        self.tracksToAdd.append(trackDict)
        self.trackIdlist.append(tId)

    def add_maf(self, data, trackData):
        """

        self.tracksToAdd.append(trackDict)
        self.trackIdlist.append(tId)
        os.unlink(gff3)

    def add_bigwig(self, data, trackData):
        useuri = trackData["useuri"].lower() == "yes"
        if useuri:
            url = data
        else:
            url = "%s.bigwig" % trackData["label"]

            # Return non-human label for use in other fields
            yield outputTrackConfig["label"]

    def add_default_session(self, default_data):
        """
        default session settings are hard and fragile.
        .add_default_view() and other configuration code adapted from
         https://github.com/abretaud/tools-iuc/blob/jbrowse2/tools/jbrowse2/jbrowse2.py
        """
        tracks_data = []
        # TODO using the default session for now, but check out session specs in the future https://github.com/GMOD/jbrowse-components/issues/2708
        track_types = {}
        with open(self.config_json_file, "r") as config_file:
            config_json = json.load(config_file)
        if self.config_json:
            config_json.update(self.config_json)
        for track_conf in self.tracksToAdd:
            tId = track_conf["trackId"]
            track_types[tId] = track_conf["type"]
            style_data = default_data["style"].get(tId,  None)
            if not style_data:
                logging.warn("### No style data in default data %s for %s" % (default_data, tId))
                style_data = {"type": "LinearBasicDisplay"}
            if "displays" in track_conf:
                disp = track_conf["displays"][0]["type"]
                style_data["type"] = disp
            if track_conf.get("style_labels", None):

                    "displays": [style_data],
                }
            )
        # The view for the assembly we're adding
        view_json = {"type": "LinearGenomeView", "tracks": tracks_data}
        logging.warn("### view_json=%s" % view_json)
        refName = None
        drdict = {
            "reversed": False,
            "assemblyName": self.genome_name,
            "start": 1,

        config_json["configuration"].update(config_data)
        self.config_json.update(config_json)
        with open(config_path, "w") as config_file:
            json.dump(self.config_json, config_file, indent=2)

    def clone_jbrowse(self, realclone=False):
        """Clone a JBrowse directory into a destination directory. This also works in Biocontainer testing now
        Leave as True between version updates on temporary tools - requires manual conda trigger :(
        """
        dest = self.outdir
        if realclone:
            self.subprocess_check_call(
                ["jbrowse", "create", dest, "-f", "--tag", f"{JB2VER}"]
            )

            for x in root.findall("metadata/genomes/genome")
        ],
    )
    jc.process_genomes()

    default_session_data = {
        "visibility": {
            "default_on": [],
            "default_off": [],
        },

        trackfiles = track.findall("files/trackFile")
        if trackfiles:
            for x in track.findall("files/trackFile"):
                track_conf["label"] = x.attrib["label"]
                trackkey = track_conf["label"]
                track_conf["useuri"] = x.attrib["useuri"]
                if is_multi_bigwig:
                    multi_bigwig_paths.append(
                        (
                            x.attrib["label"],

                    {},  # No metadata for multiple bigwig
                )
            )
        track_conf["category"] = track.attrib["cat"]
        track_conf["format"] = track.attrib["format"]
        track_conf["conf"] = etree_to_dict(track.find("options"))
        track_conf["category"] = track.attrib["cat"]
        track_conf["format"] = track.attrib["format"]
        keys = jc.process_annotations(track_conf)

        if keys:
            for key in keys:
                default_session_data["visibility"][
                    track.attrib.get("visibility", "default_off")
                ].append(key)
            if track.find("options/style"):
                default_session_data["style"][key] = {
                    item.tag: parse_style_conf(item) for item in track.find("options/style")
                }
                logging.warn("### added %s to defsess %s for %s" % (trackkey, default_session_data, key ))
            else:
                default_session_data["style"][key] = {}
                logging.warn("@@@@ no options/style found for %s" % (key))

            if track.find("options/style_labels"):
                default_session_data["style_labels"][key] = {
                    item.tag: parse_style_conf(item)
                    for item in track.find("options/style_labels")
                }
    default_session_data["defaultLocation"] = root.find(
        "metadata/general/defaultLocation"
    ).text
    default_session_data["session_name"] = root.find(
        "metadata/general/session_name"