The SAbDab CSV has one column heading that changes to include the month and year. This required manual updates periodically. Currently in the tests it just logs that the column is missing.
New logic will search the list of column headings to find the first one that starts with Highest_Clin_Trial. If none is found, it uses Highest_Clin_Trial which will fail but at least be shown in the error message.
Since I was making changes anyway, I updated the baseVersion of ChEMBL to the latest (31), because the current behavior is to keep trying versions until it finds one, which leads to a bunch of ftp errors. (This may be a candidate for future development trying to figure out the version from the README or something.)
The SAbDab CSV has one column heading that changes to include the month and year. This required manual updates periodically. Currently in the tests it just logs that the column is missing.
New logic will search the list of column headings to find the first one that starts with
Highest_Clin_Trial
. If none is found, it usesHighest_Clin_Trial
which will fail but at least be shown in the error message.Since I was making changes anyway, I updated the baseVersion of ChEMBL to the latest (31), because the current behavior is to keep trying versions until it finds one, which leads to a bunch of ftp errors. (This may be a candidate for future development trying to figure out the version from the README or something.)