How to analyse expression of long non-coding genes using RNA-seq

[mayankkaashyap]

Hi,
I am aiming to study long non-coding RNA in plants for which reference genome and transcriptome are available. Long non-coding RNA are reported to be <200 - 500 nt.
1) What insert size of RNA-seq reads I should be generating (using TruSeq Total RNA Illumina kit)
2) How much coverage I need for 740 Mbp genome size.

[colindaven]

What is the quality of the genome/transcriptome? Are you aiming to find novel lncRNA or just quantify existing lncRNAs ?

1) standard insert size should be fine.
2) Coverage makes sense for genome sequencing, not for RNA-seq.

lncRNAs are lowly expressed as you know, so you need to sequence very deeply.

For RNA-seq and delineation of novel lncRNAs, the more the better. Certainly not less than four/six samples per lane on an Illumina HiSEq 2500.

For transcript quantification, you will probably get away with less. (six-eight samples per lane).

If a transcriptome already exists, can you reanalyse the transcriptome to find / refind existing lncRNAs ?

[mayankkaashyap]

Hi,
Thanks for reply.
I am planning to put 6 samples/lanes on Hi-Seq 3000. Reference genome is of good quality. I have to dig out more as no lncRNAs were identified from transcriptome. Even I want to see their response at stress, so this study will be novel in its sense.
Do you mean standard as 150 bp inserts??