Age | Commit message (Collapse) | Author |
|
cofactors. For some reason the rows/columns are inverted when converted to a python object when doing scanone with cofactors vs without cofactors
|
|
same file is being written to simultaneously (or being written to while it's being read)
|
|
the results are not converted into a Python object in a way that preserves marker names (which is important because pseudomarkers can be added)
Instead the marker names are extracted from the scanone results using R immediately after they're generated, and then passed to process_rqtl_results
|
|
previously it could throw an error
|
|
collection the use set as their default collection
|
|
input string
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
instead of just checking if there are case attributes
|
|
instead check categorical_attr_exists
|
|
apparently the boolean doesn't get passed to the template properly
|
|
case attributes with fewer than 10 distinct values, since it currently throws a JS error if case attributes exist but none have fewer than 10 distinct values (specifically when we have RRID as a case attribute)
|
|
|
|
with calculating PCA that I don't know the cause of yet
|
|
many shared samples torun scipy.stats.pearsonr/spearmanr
|
|
error caused by an empty string be stored in a collection's trait list
|
|
number is needed for the links
|
|
Redis as both string and bytes
|
|
drawing the Gene Band
|
|
|
|
rat IDs don't have any prefix with a colon
|
|
situations where a case attribute also has a link
|
|
in the trait page sample table
|
|
being returned as bytes instead of a str
|
|
|
|
values are sometimes stored as bytes and sometimes as strings
|
|
groups in Redis are stored as strings
|
|
|
|
|
|
be changed for Python 3.8
|
|
|
|
|
|
newline character
|
|
resulting from a trait missing both pre_publication_description and pubmed ID
|
|
|
|
being displayed correctly
|
|
|
|
Redis.smembers apparently returns values as bytes
|
|
|