Added argument for specifying NetMHCIIpan/NetMHCIIpanEL version

pvactools/lib/netmhc_pan_version.py

@@ -0,0 +1,28 @@
+import argparse
+
+
+class NetMHCPanVersion:
+    valid_versions = ["4.3", "4.2", "4.1", "4.0 (Not supported by standalone IEDB)"]
+
+    def __init__(self, list):
+        self.list = list
+
+    def print_valid_versions(self):
+        if self.list:
+            print("Valid NetMHCIIpan and NetMHCIIpanEL Versions")
+            print('\n'.join([a for a in self.valid_versions]))
+
+    @classmethod
+    def parser(cls, tool="pvacseq"):
+        parser = argparse.ArgumentParser(
+            "%s valid_netmhcpan_versions" % tool,
+            description="Show a list of valid versions of NetMHCIIpan and NetMHCIIpanEL that can be used.",
+            formatter_class=argparse.ArgumentDefaultsHelpFormatter
+        )
+        parser.add_argument(
+            '-l', '--list',
+            help="List the valid NetMHCIIpan and NetMHCIIpanEL versions.",
+            default='None',
+            action='store_true'
+        )
+        return parser

pvactools/lib/output_parser.py

@@ -172,8 +172,11 @@ def get_scores(self, line, method):
             return {'score': float(line['BigMHC_IM'])}
         elif method.lower() == 'netmhcpan_el':
             return {'score': float(line['score'])}
-        elif method.lower() == 'netmhciipan_el':
-            return {'score': float(line['score'])}
+        elif 'netmhciipan_el' in method.lower():
+            try:
+                return {'score': float(line['score'])}
+            except:
+                return {'ic50': float(line['ic50'])}
         else:
             return {'ic50': float(line['ic50'])}
 
@@ -596,7 +599,7 @@ def output_headers(self):
                 headers.append("%s MT Score" % pretty_method)
                 continue
 
-            if method in ['netmhcpan_el', 'netmhciipan_el']:
+            if 'netmhciipan_el' in method or 'netmhcpan_el' in method:
                 headers.append("%s WT Score" % pretty_method)
                 headers.append("%s MT Score" % pretty_method)
             else:
@@ -620,9 +623,12 @@ def flurry_headers(self, headers):
 
     def prediction_methods(self):
         methods = set()
+        pattern = re.compile(rf"{re.escape(self.sample_name)}\.(\w+(?:-\d+\.\d+)?)")
+
         for input_iedb_file in self.input_iedb_files:
-            # we remove "sample_name." prefix from filename and then first part before a dot is the method name 
-            method = (os.path.basename(input_iedb_file)[len(self.sample_name)+1:]).split('.', 1)[0]
+            filename = os.path.basename(input_iedb_file)
+            match = pattern.match(filename)
+            method = match.group(1)
             methods.add(method)
 
         return sorted(list(methods))
@@ -761,6 +767,7 @@ def execute(self):
 
 
 class DefaultOutputParser(OutputParser):
+
     def parse_iedb_file(self, tsv_entries):
         with open(self.key_file, 'r') as key_file_reader:
             protein_identifiers_from_label = yaml.load(key_file_reader, Loader=yaml.FullLoader)
@@ -769,8 +776,12 @@ def parse_iedb_file(self, tsv_entries):
         for input_iedb_file in self.input_iedb_files:
             with open(input_iedb_file, 'r') as reader:
                 iedb_tsv_reader = csv.DictReader(reader, delimiter='\t')
-                # we remove "sample_name." prefix from filename and then first part before a dot is the method name 
-                method = (os.path.basename(input_iedb_file)[len(self.sample_name)+1:]).split('.', 1)[0]
+                filename = os.path.basename(input_iedb_file)
+
+                pattern = re.compile(rf"{re.escape(self.sample_name)}\.(\w+(?:-\d+\.\d+)?)")
+                match = pattern.match(filename)
+                method = match.group(1)
+
                 for line in iedb_tsv_reader:
                     if "Warning: Potential DNA sequence(s)" in line['allele']:
                         continue
@@ -831,8 +842,12 @@ def parse_iedb_file(self):
         for input_iedb_file in self.input_iedb_files:
             with open(input_iedb_file, 'r') as reader:
                 iedb_tsv_reader = csv.DictReader(reader, delimiter='\t')
-                # we remove "sample_name." prefix from filename and then first part before a dot is the method name 
-                method = (os.path.basename(input_iedb_file)[len(self.sample_name)+1:]).split('.', 1)[0]
+                filename = os.path.basename(input_iedb_file)
+
+                pattern = re.compile(rf"{re.escape(self.sample_name)}\.(\w+(?:-\d+\.\d+)?)")
+                match = pattern.match(filename)
+                method = match.group(1)
+
                 for line in iedb_tsv_reader:
                     if "Warning: Potential DNA sequence(s)" in line['allele']:
                         continue
@@ -946,7 +961,7 @@ def output_headers(self):
                 headers.append("%s Score" % pretty_method)
                 continue
 
-            if method in ['netmhcpan_el', 'netmhciipan_el']:
+            if 'netmhciipan_el' in method or 'netmhcpan_el' in method:
                 headers.append("%s Score" % pretty_method)
             else:
                 headers.append("%s IC50 Score" % pretty_method)
@@ -1017,8 +1032,12 @@ def parse_iedb_file(self):
             # input iedb file
             with open(input_iedb_file, 'r') as reader:
                 iedb_tsv_reader = csv.DictReader(reader, delimiter='\t')
-                # we remove "sample_name." prefix from filename and then first part before a dot is the method name 
-                method = (os.path.basename(input_iedb_file)[len(self.sample_name)+1:]).split('.', 1)[0]
+                filename = os.path.basename(input_iedb_file)
+
+                pattern = re.compile(rf"{re.escape(self.sample_name)}\.(\w+(?:-\d+\.\d+)?)")
+                match = pattern.match(filename)
+                method = match.group(1)
+
                 # header: allele, seq_num, start, end, length, peptide, ic50, percentile_rank
                 for line in iedb_tsv_reader:
                     if "Warning: Potential DNA sequence(s)" in line['allele']:

pvactools/lib/prediction_class.py

@@ -653,9 +653,13 @@ def url(self):
     def parse_iedb_allele_file(self):
         #Ultimately we probably want this method to call out to IEDB but their command is currently broken
         #curl --data "method=ann&species=human" http://tools-api.iedb.org/tools_api/mhci/
+        file_name = next(
+            (name for name in ["netmhciipan", "netmhciipan_el"] if name in self.iedb_prediction_method),
+            self.iedb_prediction_method
+        )
         base_dir               = os.path.abspath(os.path.join(os.path.dirname(os.path.realpath(__file__)), '..'))
         iedb_alleles_dir       = os.path.join(base_dir, 'tools', 'pvacseq', 'iedb_alleles', 'class_ii')
-        iedb_alleles_file_name = os.path.join(iedb_alleles_dir, "%s.tsv" % self.iedb_prediction_method)
+        iedb_alleles_file_name = os.path.join(iedb_alleles_dir, "%s.tsv" % file_name)
         alleles = []
         with open(iedb_alleles_file_name) as iedb_alleles_file:
             for row in iedb_alleles_file:
@@ -675,14 +679,21 @@ def iedb_executable_params(self, iedb_executable_path, method, allele, input_fil
         allele = allele.replace('-DPB', '/DPB').replace('-DQB', '/DQB')
         return [iedb_executable_path, method, allele, input_file, str(epitope_length)]
 
+class NetMHCIIVersion:
+    netmhcii_pan_version = None
+
 class NetMHCIIpan(IEDBMHCII):
     @property
     def iedb_prediction_method(self):
-        return 'NetMHCIIpan'
+        if NetMHCIIVersion.netmhcii_pan_version in ['4.0', '4.2', '4.3']:
+            return 'netmhciipan_ba-' + NetMHCIIVersion.netmhcii_pan_version
+        return 'netmhciipan_ba'
 
 class NetMHCIIpanEL(IEDBMHCII):
     @property
     def iedb_prediction_method(self):
+        if NetMHCIIVersion.netmhcii_pan_version in ['4.0', '4.2', '4.3']:
+            return 'netmhciipan_el-' + NetMHCIIVersion.netmhcii_pan_version
         return 'netmhciipan_el'
 
 class NNalign(IEDBMHCII):

pvactools/lib/run_argument_parser.py

@@ -111,6 +111,12 @@ def __init__(self, tool_name, input_file_help):
             default=1,
             help="Number of threads to use for parallelizing peptide-MHC binding prediction calls.",
         )
+        parser.add_argument(
+            "--netmhc_pan_version",
+            choices=["4.3", "4.2", "4.1", "4.0"],
+            default=None,
+            help="Specify the version of NetMHCIIpan or NetMHCIIpanEL to be used during the run.",
+        )
         self.parser = parser
 
     def prediction_args(self):

pvactools/tools/pvacbind/__init__.py

@@ -8,6 +8,7 @@
     'top_score_filter',
     'net_chop',
     'netmhc_stab',
+    'netmhc_pan_version',
     'calculate_reference_proteome_similarity',
     'generate_aggregated_report',
     'identify_problematic_amino_acids',

pvactools/tools/pvacbind/main.py

@@ -87,6 +87,13 @@ def main():
     )
     valid_algorithms_parser.set_defaults(func=valid_algorithms)
 
+    netmhc_pan_version_parser = subparsers.add_parser(
+        "valid_netmhcpan_versions",
+        help="Show a list of valid versions of NetMHCIIpan and NetMHCIIpanEL that can be used.",
+        add_help=False
+    )
+    netmhc_pan_version_parser.set_defaults(func=netmhc_pan_version)
+
     allele_specific_cutoffs_parser = subparsers.add_parser(
         "allele_specific_cutoffs",
         help="Show the allele specific cutoffs",

pvactools/tools/pvacbind/netmhc_pan_version.py

@@ -0,0 +1,15 @@
+import sys
+
+from pvactools.lib.netmhc_pan_version import NetMHCPanVersion
+
+def define_parser():
+    return NetMHCPanVersion.parser('pvacbind')
+
+def main(args_input = sys.argv[1:]):
+    parser = define_parser()
+    args = parser.parse_args(args_input)
+
+    NetMHCPanVersion(args.list).print_valid_versions()
+
+if __name__ == "__main__":
+    main()

pvactools/tools/pvacfuse/__init__.py

@@ -9,6 +9,7 @@
     'top_score_filter',
     'net_chop',
     'netmhc_stab',
+    'netmhc_pan_version',
     'calculate_reference_proteome_similarity',
     'generate_protein_fasta',
     "generate_aggregated_report",

pvactools/tools/pvacfuse/main.py

@@ -85,6 +85,13 @@ def define_parser():
     )
     valid_algorithms_parser.set_defaults(func=valid_algorithms)
 
+    netmhc_pan_version_parser = subparsers.add_parser(
+        "valid_netmhcpan_versions",
+        help="Show a list of valid versions of NetMHCIIpan and NetMHCIIpanEL that can be used.",
+        add_help=False
+    )
+    netmhc_pan_version_parser.set_defaults(func=netmhc_pan_version)
+
     identify_problematic_amino_acids_parser = subparsers.add_parser(
         "identify_problematic_amino_acids",
         help="Mark problematic amino acid positions in each epitope or filter entries that have problematic amino acids.",

pvactools/tools/pvacfuse/netmhc_pan_version.py

@@ -0,0 +1,15 @@
+import sys
+
+from pvactools.lib.netmhc_pan_version import NetMHCPanVersion
+
+def define_parser():
+    return NetMHCPanVersion.parser('pvacfuse')
+
+def main(args_input = sys.argv[1:]):
+    parser = define_parser()
+    args = parser.parse_args(args_input)
+
+    NetMHCPanVersion(args.list).print_valid_versions()
+
+if __name__ == "__main__":
+    main()

pvactools/tools/pvacseq/__init__.py

@@ -12,6 +12,7 @@
     'top_score_filter',
     'net_chop',
     'netmhc_stab',
+    'netmhc_pan_version',
     'calculate_reference_proteome_similarity',
     'transcript_support_level_filter',
     'identify_problematic_amino_acids',

pvactools/tools/pvacseq/main.py

@@ -114,6 +114,13 @@ def define_parser():
     )
     valid_algorithms_parser.set_defaults(func=valid_algorithms)
 
+    netmhc_pan_version_parser = subparsers.add_parser(
+        "valid_netmhcpan_versions",
+        help="Show a list of valid versions of NetMHCIIpan and NetMHCIIpanEL that can be used.",
+        add_help=False
+    )
+    netmhc_pan_version_parser.set_defaults(func=netmhc_pan_version)
+
     allele_specific_cutoffs_parser = subparsers.add_parser(
         "allele_specific_cutoffs",
         help="Show the allele specific cutoffs.",

pvactools/tools/pvacseq/netmhc_pan_version.py

@@ -0,0 +1,15 @@
+import sys
+
+from pvactools.lib.netmhc_pan_version import NetMHCPanVersion
+
+def define_parser():
+    return NetMHCPanVersion.parser('pvacseq')
+
+def main(args_input = sys.argv[1:]):
+    parser = define_parser()
+    args = parser.parse_args(args_input)
+
+    NetMHCPanVersion(args.list).print_valid_versions()
+
+if __name__ == "__main__":
+    main()

pvactools/tools/pvacseq/run.py

@@ -74,6 +74,14 @@ def main(args_input = sys.argv[1:]):
     input_file_type = 'vcf'
     base_output_dir = os.path.abspath(args.output_dir)
 
+    if (args.netmhc_pan_version == '4.0' and args.iedb_install_directory is not None):
+        raise Exception("Standalone IEDB does not support version 4.0")
+
+    if (args.netmhc_pan_version and ("NetMHCIIpan" in args.prediction_algorithms or "NetMHCIIpanEL" in args.prediction_algorithms)):
+        NetMHCIIVersion.netmhcii_pan_version = args.netmhc_pan_version
+    elif (args.netmhc_pan_version and not ("NetMHCIIpan" in args.prediction_algorithms or "NetMHCIIpanEL" in args.prediction_algorithms or "all_class_ii" in args.prediction_algorithms)):
+        raise Exception("NetMHCIIpan and NetMHCIIpanEL version was specified but neither algorithm was selected")
+
     (class_i_prediction_algorithms, class_ii_prediction_algorithms) = split_algorithms(args.prediction_algorithms)
     alleles = combine_class_ii_alleles(args.allele)
     (class_i_alleles, class_ii_alleles, species) = split_alleles(alleles)

pvactools/tools/pvacsplice/__init__.py

@@ -9,6 +9,7 @@
     'identify_problematic_amino_acids',
     'net_chop',
     'netmhc_stab',
+    'netmhc_pan_version',
     'run',
     'top_score_filter',
     'transcript_support_level_filter',

pvactools/tools/pvacsplice/main.py

@@ -100,6 +100,13 @@ def define_parser():
     )
     valid_alleles_parser.set_defaults(func=valid_alleles)
 
+    netmhc_pan_version_parser = subparsers.add_parser(
+        "valid_netmhcpan_versions",
+        help="Show a list of valid versions of NetMHCIIpan and NetMHCIIpanEL that can be used.",
+        add_help=False
+    )
+    netmhc_pan_version_parser.set_defaults(func=netmhc_pan_version)
+
     allele_specific_cutoffs_parser = subparsers.add_parser(
         "allele_specific_cutoffs",
         help="Show the allele specific cutoffs.",

pvactools/tools/pvacsplice/netmhc_pan_version.py

@@ -0,0 +1,15 @@
+import sys
+
+from pvactools.lib.netmhc_pan_version import NetMHCPanVersion
+
+def define_parser():
+    return NetMHCPanVersion.parser('pvacsplice')
+
+def main(args_input = sys.argv[1:]):
+    parser = define_parser()
+    args = parser.parse_args(args_input)
+
+    NetMHCPanVersion(args.list).print_valid_versions()
+
+if __name__ == "__main__":
+    main()

pvactools/tools/pvacvector/__init__.py

@@ -3,6 +3,7 @@
     'visualize',
     'valid_alleles',
     'valid_algorithms',
+    'netmhc_pan_version',
     'allele_specific_cutoffs',
     'download_example_data',
 ]

pvactools/tools/pvacvector/main.py

@@ -35,6 +35,13 @@ def define_parser():
     )
     valid_algorithms_parser.set_defaults(func=valid_algorithms)
 
+    netmhc_pan_version_parser = subparsers.add_parser(
+        "valid_netmhcpan_versions",
+        help="Show a list of valid versions of NetMHCIIpan and NetMHCIIpanEL that can be used.",
+        add_help=False
+    )
+    netmhc_pan_version_parser.set_defaults(func=netmhc_pan_version)
+
     allele_specific_cutoffs_parser = subparsers.add_parser(
         "allele_specific_cutoffs",
         help="Show the allele specific cutoffs",

pvactools/tools/pvacvector/netmhc_pan_version.py

@@ -0,0 +1,15 @@
+import sys
+
+from pvactools.lib.netmhc_pan_version import NetMHCPanVersion
+
+def define_parser():
+    return NetMHCPanVersion.parser('pvacvector')
+
+def main(args_input = sys.argv[1:]):
+    parser = define_parser()
+    args = parser.parse_args(args_input)
+
+    NetMHCPanVersion(args.list).print_valid_versions()
+
+if __name__ == "__main__":
+    main()