Fix cases that MiSeq samplesheet misses index or index2

VERSIONLOG.md

@@ -1,5 +1,9 @@
 # TACA Version Log
 
+## 20240315.1
+
+Fix cases that MiSeq samplesheet misses index or index2
+
 ## 20240304.1
 
 - Make sure TACA can handle runs that generate NO sequencing data at all

taca/analysis/analysis.py

@@ -115,7 +115,7 @@ def _upload_to_statusdb(run):
     # Check if I have NoIndex lanes
     for element in parser.obj["samplesheet_csv"]:
         if (
-            "NoIndex" in element["index"] or not element["index"]
+            "NoIndex" in element.get("index", "") or not element.get("index")
         ):  # NoIndex in the case of HiSeq, empty in the case of HiSeqX
             lane = element["Lane"]  # This is a lane with NoIndex
             # In this case PF Cluster is the number of undetermined reads

taca/illumina/Runs.py

@@ -453,6 +453,10 @@ def _classify_lanes(self, samplesheets):
         # Prepare a list for lanes with NoIndex samples
         noindex_lanes = []
         for entry in self.runParserObj.samplesheet.data:
+            if not entry.get("index"):
+                entry["index"] = ""
+            if not entry.get("index2"):
+                entry["index2"] = ""
             if entry["index"].upper() == "NOINDEX" or (
                 entry["index"] == "" and entry["index2"] == ""
             ):

taca/illumina/Standard_Runs.py

@@ -131,6 +131,10 @@ def _classify_samples(self, indexfile, ssparser, runSetup):
             sample_name = sample.get("Sample_Name") or sample.get("SampleName")
             umi_length = [0, 0]
             read_length = read_cycles
+            if not sample.get("index"):
+                sample["index"] = ""
+            if not sample.get("index2"):
+                sample["index2"] = ""
             # Read the length of read 1 and read 2 from the field Recipe
             if sample.get("Recipe") and RECIPE_PAT.findall(sample.get("Recipe")):
                 ss_read_length = [