diff jbrowse2.py @ 19:bde6b1d09f7d draft

planemo upload for repository https://github.com/usegalaxy-eu/temporary-tools/tree/master/jbrowse2 commit 1290bf486bc55c02fecd0327de10a28655a18e81-dirty

--- a/jbrowse2.py	Mon Jan 29 02:34:43 2024 +0000
+++ b/jbrowse2.py	Tue Jan 30 06:05:03 2024 +0000
@@ -3,7 +3,6 @@
 import argparse
 import binascii
 import datetime
-import hashlib
 import json
 import logging
 import os
@@ -23,7 +22,7 @@
 
 TODAY = datetime.datetime.now().strftime("%Y-%m-%d")
 GALAXY_INFRASTRUCTURE_URL = None
-JB2REL = "v2.10.0"
+JB2REL = "v2.10.1"
 # version pinned for cloning
 
 mapped_chars = {
@@ -232,7 +231,9 @@
         elif "scaling" in track:
             if track["scaling"]["method"] == "ignore":
                 if track["scaling"]["scheme"]["color"] != "__auto__":
-                    trackConfig["style"]["color"] = track["scaling"]["scheme"]["color"]
+                    trackConfig["style"]["color"] = track["scaling"]["scheme"][
+                        "color"
+                    ]
                 else:
                     trackConfig["style"]["color"] = self.hex_from_rgb(
                         *self._get_colours()
@@ -259,13 +260,18 @@
                             "blue": blue,
                         }
                     )
-                    trackConfig["style"]["color"] = color_function.replace("\n", "")
+                    trackConfig["style"]["color"] = color_function.replace(
+                        "\n", ""
+                    )
                 elif trackFormat == "gene_calls":
                     # Default values, based on GFF3 spec
                     min_val = 0
                     max_val = 1000
                     # Get min/max and build a scoring function since JBrowse doesn't
-                    if scales["type"] == "automatic" or scales["type"] == "__auto__":
+                    if (
+                        scales["type"] == "automatic"
+                        or scales["type"] == "__auto__"
+                    ):
                         min_val, max_val = self.min_max_gff(gff3)
                     else:
                         min_val = scales.get("min", 0)
@@ -273,7 +279,9 @@
 
                     if scheme["color"] == "__auto__":
                         user_color = "undefined"
-                        auto_color = "'%s'" % self.hex_from_rgb(*self._get_colours())
+                        auto_color = "'%s'" % self.hex_from_rgb(
+                            *self._get_colours()
+                        )
                     elif scheme["color"].startswith("#"):
                         user_color = "'%s'" % self.hex_from_rgb(
                             *self.rgb_from_hex(scheme["color"][1:])
@@ -281,7 +289,9 @@
                         auto_color = "undefined"
                     else:
                         user_color = "undefined"
-                        auto_color = "'%s'" % self.hex_from_rgb(*self._get_colours())
+                        auto_color = "'%s'" % self.hex_from_rgb(
+                            *self._get_colours()
+                        )
 
                     color_function = self.COLOR_FUNCTION_TEMPLATE_QUAL.format(
                         **{
@@ -293,7 +303,9 @@
                         }
                     )
 
-                    trackConfig["style"]["color"] = color_function.replace("\n", "")
+                    trackConfig["style"]["color"] = color_function.replace(
+                        "\n", ""
+                    )
         return trackConfig
 
 
@@ -336,40 +348,41 @@
     for (key, value) in node.findall("dataset")[0].attrib.items():
         metadata["dataset_%s" % key] = value
 
-    for (key, value) in node.findall("history")[0].attrib.items():
-        metadata["history_%s" % key] = value
-
-    for (key, value) in node.findall("metadata")[0].attrib.items():
-        metadata["metadata_%s" % key] = value
-
-    for (key, value) in node.findall("tool")[0].attrib.items():
-        metadata["tool_%s" % key] = value
+    if node.findall("history"):
+        for (key, value) in node.findall("history")[0].attrib.items():
+            metadata["history_%s" % key] = value
 
-    # Additional Mappings applied:
-    metadata[
-        "dataset_edam_format"
-    ] = '<a target="_blank" href="http://edamontology.org/{0}">{1}</a>'.format(
-        metadata["dataset_edam_format"], metadata["dataset_file_ext"]
-    )
-    metadata["history_user_email"] = '<a href="mailto:{0}">{0}</a>'.format(
-        metadata["history_user_email"]
-    )
-    metadata["hist_name"] = metadata["history_display_name"]
-    metadata[
-        "history_display_name"
-    ] = '<a target="_blank" href="{galaxy}/history/view/{encoded_hist_id}">{hist_name}</a>'.format(
-        galaxy=GALAXY_INFRASTRUCTURE_URL,
-        encoded_hist_id=metadata["history_id"],
-        hist_name=metadata["history_display_name"],
-    )
-    metadata[
-        "tool_tool"
-    ] = '<a target="_blank" href="{galaxy}/datasets/{encoded_id}/show_params">{tool_id}</a>'.format(
-        galaxy=GALAXY_INFRASTRUCTURE_URL,
-        encoded_id=metadata["dataset_id"],
-        tool_id=metadata["tool_tool_id"],
-        # tool_version=metadata['tool_tool_version'],
-    )
+    if node.findall("metadata"):
+        for (key, value) in node.findall("metadata")[0].attrib.items():
+            metadata["metadata_%s" % key] = value
+            # Additional Mappings applied:
+            metadata[
+                "dataset_edam_format"
+            ] = '<a target="_blank" href="http://edamontology.org/{0}">{1}</a>'.format(
+                metadata["dataset_edam_format"], metadata["dataset_file_ext"]
+            )
+            metadata["history_user_email"] = '<a href="mailto:{0}">{0}</a>'.format(
+                metadata["history_user_email"]
+            )
+            metadata["hist_name"] = metadata["history_display_name"]
+            metadata[
+                "history_display_name"
+            ] = '<a target="_blank" href="{galaxy}/history/view/{encoded_hist_id}">{hist_name}</a>'.format(
+                galaxy=GALAXY_INFRASTRUCTURE_URL,
+                encoded_hist_id=metadata["history_id"],
+                hist_name=metadata["history_display_name"],
+            )
+    if node.findall("tool"):
+        for (key, value) in node.findall("tool")[0].attrib.items():
+            metadata["tool_%s" % key] = value
+        metadata[
+            "tool_tool"
+        ] = '<a target="_blank" href="{galaxy}/datasets/{encoded_id}/show_params">{tool_id}{tool_version}</a>'.format(
+            galaxy=GALAXY_INFRASTRUCTURE_URL,
+            encoded_id=metadata.get("dataset_id", ""),
+            tool_id=metadata.get("tool_tool_id", ""),
+            tool_version=metadata.get("tool_tool_version",""),
+        )
     return metadata
 
 
@@ -389,7 +402,9 @@
 
     def subprocess_check_call(self, command, output=None):
         if output:
-            log.debug("cd %s && %s >  %s", self.outdir, " ".join(command), output)
+            log.debug(
+                "cd %s && %s >  %s", self.outdir, " ".join(command), output
+            )
             subprocess.check_call(command, cwd=self.outdir, stdout=output)
         else:
             log.debug("cd %s && %s", self.outdir, " ".join(command))
@@ -468,13 +483,8 @@
             self.config_json["assemblies"] = assemblies
 
     def make_assembly(self, fapath, gname):
-        hashData = [
-            fapath,
-            gname,
-        ]
-        hashData = "|".join(hashData).encode("utf-8")
-        ghash = hashlib.md5(hashData).hexdigest()
-        faname = ghash + ".fa.gz"
+
+        faname = gname + ".fa.gz"
         fadest = os.path.join(self.outdir, faname)
         cmd = "bgzip -i -c %s -I %s.gzi > %s && samtools faidx %s" % (
             fapath,
@@ -495,6 +505,7 @@
                 "uri": faname + ".gzi",
             },
         }
+        self.genome_sequence_adapter = adapter
         trackDict = {
             "name": gname,
             "sequence": {
@@ -604,7 +615,7 @@
             "plugins": [
                 {
                     "name": "MafViewer",
-                    "url": "https://unpkg.com/browse/jbrowse-plugin-mafviewer@1.0.6/dist/jbrowse-plugin-mafviewer.umd.production.min.js",
+                    "url": "https://unpkg.com/jbrowse-plugin-mafviewer/dist/jbrowse-plugin-mafviewer.umd.production.min.js"
                 }
             ]
         }
@@ -623,7 +634,9 @@
         self.subprocess_check_call(cmd)
         # Construct samples list
         # We could get this from galaxy metadata, not sure how easily.
-        ps = subprocess.Popen(["grep", "^s [^ ]*", "-o", data], stdout=subprocess.PIPE)
+        ps = subprocess.Popen(
+            ["grep", "^s [^ ]*", "-o", data], stdout=subprocess.PIPE
+        )
         output = subprocess.check_output(("sort", "-u"), stdin=ps.stdout)
         ps.wait()
         outp = output.decode("ascii")
@@ -783,7 +796,9 @@
         url = fname
         self.subprocess_check_call(["cp", data, dest])
         bloc = {"uri": url}
-        if bam_index is not None and os.path.exists(os.path.realpath(bam_index)):
+        if bam_index is not None and os.path.exists(
+            os.path.realpath(bam_index)
+        ):
             # bai most probably made by galaxy and stored in galaxy dirs, need to copy it to dest
             self.subprocess_check_call(
                 ["cp", os.path.realpath(bam_index), dest + ".bai"]
@@ -794,7 +809,9 @@
             #      => no index generated by galaxy, but there might be one next to the symlink target
             #      this trick allows to skip the bam sorting made by galaxy if already done outside
             if os.path.exists(os.path.realpath(data) + ".bai"):
-                self.symlink_or_copy(os.path.realpath(data) + ".bai", dest + ".bai")
+                self.symlink_or_copy(
+                    os.path.realpath(data) + ".bai", dest + ".bai"
+                )
             else:
                 log.warn("Could not find a bam index (.bai file) for %s", data)
         trackDict = {
@@ -823,12 +840,62 @@
         self.tracksToAdd.append(trackDict)
         self.trackIdlist.append(tId)
 
+    def add_cram(self, data, trackData, cramOpts, cram_index=None, **kwargs):
+        tId = trackData["label"]
+        fname = "%s.cram" % trackData["label"]
+        dest = "%s/%s" % (self.outdir, fname)
+        url = fname
+        self.subprocess_check_call(["cp", data, dest])
+        bloc = {"uri": url}
+        if cram_index is not None and os.path.exists(
+            os.path.realpath(cram_index)
+        ):
+            # most probably made by galaxy and stored in galaxy dirs, need to copy it to dest
+            self.subprocess_check_call(
+                ["cp", os.path.realpath(cram_index), dest + ".crai"]
+            )
+        else:
+            # Can happen in exotic condition
+            # e.g. if bam imported as symlink with datatype=unsorted.bam, then datatype changed to bam
+            #      => no index generated by galaxy, but there might be one next to the symlink target
+            #      this trick allows to skip the bam sorting made by galaxy if already done outside
+            if os.path.exists(os.path.realpath(data) + ".crai"):
+                self.symlink_or_copy(
+                    os.path.realpath(data) + ".crai", dest + ".crai"
+                )
+            else:
+                log.warn(
+                    "Could not find a cram index (.crai file) for %s", data
+                )
+        trackDict = {
+            "type": "AlignmentsTrack",
+            "trackId": tId,
+            "name": trackData["name"],
+            "assemblyNames": [self.genome_name],
+            "adapter": {
+                "type": "CramAdapter",
+                "cramLocation": bloc,
+                "craiLocation": {"uri": fname + ".crai",},
+                "sequenceAdapter": self.genome_sequence_adapter,
+                },
+            "displays": [
+                {
+                    "type": "LinearAlignmentsDisplay",
+                    "displayId": "%s-LinearAlignmentsDisplay" % tId,
+                },
+            ],
+        }
+        style_json = self._prepare_track_style(trackDict)
+        trackDict["style"] = style_json
+        self.tracksToAdd.append(trackDict)
+        self.trackIdlist.append(tId)
+
     def add_vcf(self, data, trackData):
         tId = trackData["label"]
-        url = "%s/api/datasets/%s/display" % (
-            self.giURL,
-            trackData["metadata"]["dataset_id"],
-        )
+        # url = "%s/api/datasets/%s/display" % (
+        # self.giURL,
+        # trackData["metadata"]["dataset_id"],
+        # )
         url = "%s.vcf.gz" % tId
         dest = "%s/%s" % (self.outdir, url)
         cmd = "bgzip -c %s  > %s" % (data, dest)
@@ -879,7 +946,9 @@
                 dest,
             )  # "gff3sort.pl --precise '%s' | grep -v \"^$\" > '%s'"
             self.subprocess_popen(cmd)
-            self.subprocess_check_call(["tabix", "-f", "-p", "gff", dest + ".gz"])
+            self.subprocess_check_call(
+                ["tabix", "-f", "-p", "gff", dest + ".gz"]
+            )
 
     def _sort_bed(self, data, dest):
         # Only index if not already done
@@ -1085,28 +1154,12 @@
 
             outputTrackConfig["key"] = track_human_label
 
-            # We add extra data to hash for the case of REST + SPARQL.
-            if (
-                "conf" in track
-                and "options" in track["conf"]
-                and "url" in track["conf"]["options"]
-            ):
-                rest_url = track["conf"]["options"]["url"]
-            else:
-                rest_url = ""
             outputTrackConfig["trackset"] = track.get("trackset", {})
-            # I chose to use track['category'] instead of 'category' here. This
-            # is intentional. This way re-running the tool on a different date
-            # will not generate different hashes and make comparison of outputs
-            # much simpler.
-            hashData = [
-                str(dataset_path),
+            outputTrackConfig["label"] = "%s_%i_%s" % (
+                dataset_ext,
+                i,
                 track_human_label,
-                track["category"],
-                rest_url,
-            ]
-            hashData = "|".join(hashData).encode("utf-8")
-            outputTrackConfig["label"] = hashlib.md5(hashData).hexdigest() + "_%s" % i
+            )
             outputTrackConfig["metadata"] = extra_metadata
             outputTrackConfig["name"] = track_human_label
 
@@ -1138,17 +1191,10 @@
                     outputTrackConfig,
                 )
             elif dataset_ext == "bam":
-                real_indexes = track["conf"]["options"]["pileup"]["bam_indices"][
-                    "bam_index"
-                ]
+                real_indexes = track["conf"]["options"]["pileup"][
+                    "bam_indices"
+                ]["bam_index"]
                 if not isinstance(real_indexes, list):
-                    # <bam_indices>
-                    #  <bam_index>/path/to/a.bam.bai</bam_index>
-                    # </bam_indices>
-                    #
-                    # The above will result in the 'bam_index' key containing a
-                    # string. If there are two or more indices, the container
-                    # becomes a list. Fun!
                     real_indexes = [real_indexes]
 
                 self.add_bam(
@@ -1157,6 +1203,19 @@
                     track["conf"]["options"]["pileup"],
                     bam_index=real_indexes[i],
                 )
+            elif dataset_ext == "cram":
+                real_indexes = track["conf"]["options"]["cram"][
+                    "cram_indices"
+                ]["cram_index"]
+                if not isinstance(real_indexes, list):
+                    real_indexes = [real_indexes]
+
+                self.add_cram(
+                    dataset_path,
+                    outputTrackConfig,
+                    track["conf"]["options"]["cram"],
+                    cram_index=real_indexes[i],
+                )
             elif dataset_ext == "blastxml":
                 self.add_blastxml(
                     dataset_path,
@@ -1290,14 +1349,18 @@
             config_json.update(self.config_json)
         config_data = {}
 
-        config_data["disableAnalytics"] = data.get("analytics", "false") == "true"
+        config_data["disableAnalytics"] = (
+            data.get("analytics", "false") == "true"
+        )
 
         config_data["theme"] = {
             "palette": {
                 "primary": {"main": data.get("primary_color", "#0D233F")},
                 "secondary": {"main": data.get("secondary_color", "#721E63")},
                 "tertiary": {"main": data.get("tertiary_color", "#135560")},
-                "quaternary": {"main": data.get("quaternary_color", "#FFB11D")},
+                "quaternary": {
+                    "main": data.get("quaternary_color", "#FFB11D")
+                },
             },
             "typography": {"fontSize": int(data.get("font_size", 10))},
         }
@@ -1351,9 +1414,10 @@
     parser = argparse.ArgumentParser(description="", epilog="")
     parser.add_argument("--xml", help="Track Configuration")
     parser.add_argument("--outdir", help="Output directory", default="out")
-    parser.add_argument("--version", "-V", action="version", version="%(prog)s 2.0.1")
+    parser.add_argument(
+        "--version", "-V", action="version", version="%(prog)s 2.0.1"
+    )
     args = parser.parse_args()
-
     tree = ET.parse(args.xml)
     root = tree.getroot()
 
@@ -1448,7 +1512,8 @@
         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")
+                item.tag: parse_style_conf(item)
+                for item in track.find("options/style")
             }
         if track.find("options/style_labels"):
             track_conf["style_labels"] = {
@@ -1461,7 +1526,9 @@
         track_conf["format"] = track.attrib["format"]
         try:
             # Only pertains to gff3 + blastxml. TODO?
-            track_conf["style"] = {t.tag: t.text for t in track.find("options/style")}
+            track_conf["style"] = {
+                t.tag: t.text for t in track.find("options/style")
+            }
         except TypeError:
             track_conf["style"] = {}
             pass
@@ -1492,7 +1559,9 @@
         "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,
+        "quaternary_color": root.find(
+            "metadata/general/quaternary_color"
+        ).text,
         "font_size": root.find("metadata/general/font_size").text,
     }
     jc.add_general_configuration(general_data)