Library type backpopulate update (#992)

* Update vcgs fastq filename regex pattern

* Add TWIST library type for some cases

* Lint

scripts/back_populate_library_type.py

@@ -44,8 +44,8 @@
 # VCGS fastq
 # This is the current name format
 vcgs_fastq_regex = (
-    r'(?P<run_date>\d{6})_(?P<g2>[A-Z\d]+)_(?P<g3>\d{4})_'
-    r'(?P<g4>[A-Z]{2}\d+)_(?P<sample_id>[\w\d-]+)_(?P<library_id>[A-Z\d-]+)_'
+    r'(?P<run_date>\d{6,8})_(?P<g2>[A-Z\d]+)_(?P<g3>\d{4})_'
+    r'(?P<g4>[A-Z]{2}\d+)_(?P<sample_id>[\w\d-]+)_(?P<library_id>[A-Z\d-]+\D?)_'
     r'(?P<library_type>[\w\d]+)_(?P<lane>L\d+)_(?P<read>R[12])\.fastq\.gz'
 )
 # Pre mid 2018 the library id was not included:
@@ -115,6 +115,12 @@ def check_assay_meta_fields(assays: list[dict], update_sequencing_groups: bool):
                 'design_description'
             )
 
+        # Some fastq files are not named in the standard format, but contain the word 'TWIST'
+        # which we can use to label then with the generic 'TWIST' library type from VCGS
+        elif 'TWIST' in fastq_filename:
+            assay_meta_fields_to_update['facility'] = 'vcgs'
+            assay_meta_fields_to_update['library_type'] = 'TWIST'
+
         else:
             logging.warning(
                 f'No file name match found for assay {assay_id}. Skipping {fastq_filename}.'