stringtie sample.bam -e -G genome.gtf -o sample.gtf
Does this line of code only output transcripts that have read support in the bam file? This results in inconsistent numbers of transcripts in the gtf files of different samples. Does the output file not include all the transcipts in the genome.gtf file (even without read support)? In subsequent analysis, prepDE.py prompted a Key error when using these inconsistent transcript gtf files. And genes that are not expressed in the first sample but are expressed in the second and subsequent samples will be lost.