For me the pipeline broke because the sample names were numeric, which the fread function then did not recognize as a column name. Since the pipeline only works if column headers are present, this implies header=TRUE.
A second concern is the all.x=TRUE but all.y=FALSE (the default). This would imply that transcripts cannot be added with later files that have not been identified in the first files. This may not be a thing because of how hisat2 etc work, but then do not set all.x either.
smoe
commented
2 years ago
This is a continuation of https://github.com/BIMSBbioinfo/pigx_rnaseq/pull/113 .
For me the pipeline broke because the sample names were numeric, which the fread function then did not recognize as a column name. Since the pipeline only works if column headers are present, this implies header=TRUE.
A second concern is the all.x=TRUE but all.y=FALSE (the default). This would imply that transcripts cannot be added with later files that have not been identified in the first files. This may not be a thing because of how hisat2 etc work, but then do not set all.x either.