We are interested in using PacBio (targeted for transcriptome sequencing) in conjunction with HiSeq/MiSeq reads to assemble (de novo) a reference transcriptome for a large non-model eukaryote.

I am looking for some information on the best approaches both in terms of library preparation but more importantly on estimating the amount of sequence needed (from PacBio). We will be running about four libraries per HiSeq lane and one library per MiSeq lane to assist with error correction.