Initial script to synchronize BII datasets with Galaxy

scde_deploy/bcbio/isatab/parser.py

@@ -123,7 +123,7 @@ def _parse_keyvals(self, line_iter):
         out = None
         line = None
         for line in line_iter:
-            if len(line) == 1:
+            if len(line) == 1 and line[0].upper() == line[0]:
                 break
             else:
                 # setup output dictionaries, trimming off blank columns

scde_deploy/scripts/bii_datasets_to_galaxy.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python
+"""Share BII data files with Galaxy using data libraries.
+
+Usage:
+  bii_datasets_to_galaxy.py <bii data dir>
+"""
+import os
+import sys
+import glob
+import collections
+
+from bcbio import isatab
+
+def main(bii_dir):
+    for study_dirs in bii_studies(bii_dir):
+        study_datafiles(study_dirs["isatab"])
+
+def study_datafiles(isatab_dir):
+    """Retrieve data files and associated metadata for a study.
+    """
+    rec = isatab.parse(isatab_dir)
+    assert len(rec.studies) == 1
+    study = rec.studies[0]
+    for assay in study.assays:
+        info = _get_assay_info(assay)
+        print info
+
+def _get_assay_info(assay):
+    """Retrieve all data files and samples associated with the assay.
+    """
+    out = collections.defaultdict(set)
+    attrs = ["Raw Data File", "Derived Data File", "Sample Name"]
+    for data_file, node in assay.nodes.items():
+        for attr in attrs:
+            for val in node.metadata.get(attr, []):
+                if val:
+                    out[attr].add(val)
+    final = {}
+    for k, v in out.iteritems():
+        final[k] = list(v)
+    return final
+
+def bii_studies(bii_dir):
+    """Retrieve meta data and raw data directories in BII.
+    """
+    bii_dirs = filter(os.path.isdir,
+                      [os.path.join(bii_dir, d) for d in os.listdir(bii_dir)])
+    metadata_dir = [d for d in bii_dirs if os.path.basename(d) == "meta_data"][0]
+    bii_dirs.remove(metadata_dir)
+    for study_dir in sorted(os.path.join(metadata_dir, d)
+                            for d in os.listdir(metadata_dir)):
+        study_name = os.path.basename(study_dir)
+        data_dirs = filter(os.path.isdir,
+                           [os.path.join(d, study_name) for d in bii_dirs])
+        assert len(data_dirs) > 0, study_name
+        yield dict(isatab=study_dir, data=data_dirs)
+
+if __name__ == "__main__":
+    main(*sys.argv[1:])