Could you please tell me the proper way in which the read names need to be formatted in a bam file containing paired end reads, for the cufflinks to recognize them as paired end reads and not single reads? Do they need to have a proper suffix (e.g. _1/_2) or the names need to be the same?
On a similar note, can cufflinks handle bam files with mixed single and paired reads?
Tnx!
It depends whether the read maps to one or multiple locations - if the aligner can determine the position of one read in a pair uniquely, but can't map the other one, I don't see why there would be a problem in using the one that mapped, especially since the reads are pretty long nowdays (>70bp).