Hello,

Could someone please clarify what the fragmentSize option really means? From the manual under the dba function it states "numeric with mean fragment size. Reads will be extended to this length before counting overlaps." but under the dba.count function it states "This value will be used as the length of the reads. Each read will be extended from its endpoint along the appropriate strand by this many bases."

Will it extend the reads TO (i.e., fragmentSize is the actual mean fragment size) or BY (i.e., actual mean fragment length is sequencing read length + fragmentSize)? If my mean actual fragment length is 200 and my sequencing read length is 75 bp, should I put the fragmentSize to 200 or 200-75=125?

Thanks in advance.

In the default case where bUseSummarizeOverlaps=TRUE when calling dba.count(), this parameter is ignored. I'll clarify in the help pages for dba() and dba.count().