How to annotate the multiploid plant using pacbio cDNA sequencing?

[huan]

I want to do the transcriptome annotiation for wheat, which is a hexapliod plant. Since there has no reference genome sequence now, we are going to sequence using pacbio RNA-seq to get the full length isoform sequences as many as possible. I have four samples and have sequenced using HiSeq 2000, then I am going to do the pacbio RNA-seq. So should I do the sequcing combined all samples together or just sequencing seperately? Is there any influence with my hexaploid wheat data analysis? BTW, should I using the IsoSeq pipline or the PacBioToCA/LSC error correction pipline?

Thanks a lot O(∩_∩)O~

[Magdoll]

This paper on wheat Iso-Seq just published btw, think you may be interested in how they did it:

http://www.biomedcentral.com/1471-2164/16/1039

They used the FLNC reads (from Iso-Seq classify) and did error correction with Illumina using proovread. I think any of the following can work: running Iso-Seq only (classify + cluster) and using Illumina for quantification and validation only, using hybrid approach (LSC or proovread)...

If you already have Iso-Seq setup I would suggest first getting the Iso-Seq pipeline run through. Then if you have additional bandwidth you can try running the hybrid approaches. I think LSC may rely pretty heavily on a good ref genome so you may want to be aware of that. Iso-Seq (and I think proovread as well) are both completely de novo.