I see that the _loh.csv file sometimes contains fewer segments than the corresponding _dnacopy.seg file. Usually it contains one or two MORE segments. I would think it would have exactly the same number of segments in the two files. In one case I have 5 segments in the _dnacopy.seg file covering most of chr18 while the _loh.csv file has only 1 segment covering only a small part of it. What is the cause of this?
I discovered a possible part of the answer. The "variants.csv" file does not have any variants within one of the segments found in the _dnacopy.seg file but not in the "loh.csv" file. Maybe the latter file omits segments not having any variants within them. The problem here is that the "num.mark" column in the "_dnacopy.seg" file has the value 6 for this segment, and I would have thought it would be 0.
I think I remember that an option was added to omit segments from the LOH file if there are no SNPs in it. However, since I'm creating that file using PureCN.R and not by calling callLOH directly, I don't think I have access to that option.