aboutsummaryrefslogtreecommitdiff
path: root/gn3/db
diff options
context:
space:
mode:
authorMuriithi Frederick Muriuki2021-09-01 07:35:40 +0300
committerMuriithi Frederick Muriuki2021-09-01 07:35:40 +0300
commitb975e0cfd1d0adc5f51e66292d29d4651d3f053f (patch)
tree9a995096f469c897b3ccb24b48c380cf27d0d4b3 /gn3/db
parent221c773daea839ecf0e50c196484bb91e3a6db33 (diff)
downloadgenenetwork3-b975e0cfd1d0adc5f51e66292d29d4651d3f053f.tar.gz
Parse the genotype file's data header
* gn3/db/genotypes.py: parse data header * tests/unit/db/test_genotypes.py: check that header's parse works correctly. Add tests to check that the parser works as expected. Add code to implement the parsing and pass the tests.
Diffstat (limited to 'gn3/db')
-rw-r--r--gn3/db/genotypes.py19
1 files changed, 19 insertions, 0 deletions
diff --git a/gn3/db/genotypes.py b/gn3/db/genotypes.py
index 2be3e1a..be0dfc2 100644
--- a/gn3/db/genotypes.py
+++ b/gn3/db/genotypes.py
@@ -87,3 +87,22 @@ def parse_genotype_labels(lines: list):
return tuple(
item for item in (__parse_label(line) for line in lines)
if item is not None)
+
+def parse_genotype_header(line: str, parlist = tuple()):
+ """
+ Parse the genotype file header line
+
+ DESCRIPTION:
+ Reworks
+ https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/utility/gen_geno_ob.py#L94-L114
+ """
+ items = [item.strip() for item in line.split("\t")]
+ Mbmap = "Mb" in items
+ prgy = ((parlist + tuple(items[4:])) if Mbmap
+ else (parlist + tuple(items[3:])))
+ return (
+ ("Mbmap", Mbmap),
+ ("cm_column", items.index("cM")),
+ ("mb_column", None if not Mbmap else items.index("Mb")),
+ ("prgy", prgy),
+ ("nprgy", len(prgy)))