Hello,
I am planning on using RNAseq with Illumina to generate a de novo transcriptome for a non-model mammal species. I'll be using a 100bp PE protocol and multiplexing 24 samples in a lane. I am curious as to what insert size will optimize downstream assembly of the transcriptome (I'm planning on using trinity for the assembly).
My PI suggested that there should be some overlap between the two read directions (e.g.: insert size of <200 with 100bp PE sequencing), but we don't know how much to shoot for. He thinks that this will eventually facilitate the de novo assembly. We just got another data set back from sequencing and it had ~10% less data than we expected, which we think was caused by too short of an insert (average <150bp), but we aren't sure.
Any suggestions are appreciated.
Thanks!
Tom
I am planning on using RNAseq with Illumina to generate a de novo transcriptome for a non-model mammal species. I'll be using a 100bp PE protocol and multiplexing 24 samples in a lane. I am curious as to what insert size will optimize downstream assembly of the transcriptome (I'm planning on using trinity for the assembly).
My PI suggested that there should be some overlap between the two read directions (e.g.: insert size of <200 with 100bp PE sequencing), but we don't know how much to shoot for. He thinks that this will eventually facilitate the de novo assembly. We just got another data set back from sequencing and it had ~10% less data than we expected, which we think was caused by too short of an insert (average <150bp), but we aren't sure.
Any suggestions are appreciated.
Thanks!
Tom