Mapping paired reads to de novo RNA-seq assemblies for quality assessment?

[BobFreemanMA]

Greetings!

I'm working with paired end reads (not strand-specific; 100 bp each end with a total fragment size of 400 bp) from Illumina v1.5 data format; 100 million read pairs, but trying out the workflow on 1m, 2m, 5m, and 10m reads. All reads are adapter trimmed and low quality base trimmed.

I'm hoping to get a few suggestions to two questions I have:

1) I am using Trinity to build de novo RNA-seq assemblies. But the mean/median assembly sizes are smaller than I would expect (503 bp/300 bp). Given the data is from mRNA from a vertebrate, what can I do to increase the mean/median contig size?

2) More importantly, as I have paired-end reads that went into the assembly, I'd like to map the pairs back onto the de novo transcripts to assess if the pairs are being used correctly, and to give me some quality assessment for each transcript. What program would folks recommend that would give me some sort of readout? I would expect something like, per transcript, X# pairs map correctly; Y# singles are mapped; Z# are mapped but violate orientation/insert size, etc.

Thanks much,
Bob

[aprendiz]

Hi BobFreemanMA

Did you ever get a reply to this?? I am trying to do exactly the same and I do not want to reinvent the wheel.

Thanks!