Conflicting if-statement

[dtaliun]

Hi Jonathon,

I think this condition will be always true when merging empirical dose files, because format_fields vector is filled out with "DS" and "LDS" a few lines above:

hds-util/src/main.cpp

Line 235 in dc0188f

if (!format_fields.empty() && emp_cnt)

Also, a related question: Is there anything specific about stats to be aware of when merging empirical doses?

Thanks and best wishes!
Daniel

[jonathonl]

Thanks for reporting this.

The one difference in the INFO fields when merging with hds-util vs imputed samples together is that the ER2 INFO field does not get recalculated (instead each sample batch ER2 is concatenated into a list). For the HDS files, ER2 cannot be recalculated because those files are missing the original genotypes. The LDS files could be recalculated, but metaMinimac doesn't use the ER2 INFO field, so it didn't seem necessary. I might change this in the future.