append mode as samplename suffix by traversing filepath

Author: keiran-rowell-unsw

multiqc/modules/proteinfold/proteinfold.py

@@ -14,6 +14,7 @@ class MultiqcModule(BaseMultiqcModule):
     The module parses results generated by a variety of protein structure prediction programs in the [ProteinFold](https://nf-co.re/proteinfold/) pipeline.
     This includes (as of release v2.0):
         - [AlphaFold2](https://github.com/google-deepmind/alphafold)
+        - [AlphaFold3](https://github.com/google-deepmind/alphafold3)
         - [ColabFold](https://github.com/sokrypton/ColabFold)
         - [ESMFold](https://github.com/facebookresearch/esm)
         - [RoseTTaFold-All-Atom](https://github.com/baker-laboratory/RoseTTAFold-All-Atom)
@@ -32,6 +33,7 @@ def __init__(self):
             href=[
                 "https://nf-co.re/proteinfold",
                 "https://github.com/google-deepmind/alphafold",
+                "https://github.com/google-deepmind/alphafold3",
                 "https://github.com/sokrypton/ColabFold",
                 "https://github.com/facebookresearch/esm",
                 "https://github.com/baker-laboratory/RoseTTAFold-All-Atom",
@@ -79,21 +81,42 @@ def __init__(self):
             "rank_19": ["_rank_19"],
             "rank_20": ["_rank_20"],
             "rank_21": ["_rank_21"],
-            "rank_22": ["_rank_22"],
+            "ranmodek_22": ["_rank_22"],
             "rank_23": ["_rank_23"],
             "rank_24": ["_rank_24"],
         }
 
+        mode_dict = {
+            "alphafold2": "AlphaFold2",
+            "alphafold3": "AlphaFold3",
+            "colabfold": "ColabFold",
+            "esmfold": "ESMFold",
+            "rosettafold2na": "RoseTTAFold2-Nucleic-Acids",
+            "rosettafold_all_atom": "RoseTTAFold-All-Atom",
+            "helixfold3": "HelixFold3",
+            "boltz": "Boltz",
+        }
+
         self.proteinfold_data: Dict[str, Any] = {}
         # I want to enable sample grouping: https://docs.seqera.io/multiqc/reports/customisation#sample-grouping
 
         for f in self.find_log_files("proteinfold"):
             self.add_data_source(f)
 
-            samplename = f["s_name"].split("_")[0]
-            clean_samplename = self.clean_s_name(samplename, f)
-            self.proteinfold_data.setdefault(clean_samplename, {})  # Set default creates if doesn't already exist
+            raw_samplename = f["s_name"].split("_")[0]
             filepath = Path(f["root"]) / f["fn"]
+            mode = "UNKNOWN"
+            for parent in filepath.parents:
+                if parent.name in mode_dict:  # traverse up the filepath until you hit a mode labelled dir
+                    mode = mode_dict[parent.name]
+                    break
+
+            mode_samplename = f"{raw_samplename}_{mode}"
+            samplename = self.clean_s_name(mode_samplename, f)
+            print(filepath)
+            print(samplename)
+
+            self.proteinfold_data.setdefault(samplename, {})  # Set default creates if doesn't already exist
 
             if f["fn"].endswith("_msa.tsv"):
                 self.proteinfold_data[samplename]["msa_depth"] = f["f"].count("\n")