I have a question for doing RNA-seq on Arabidopsis samples comparing differential gene expression between wild type and miRNA mutant lines. Just one mutant will be compared.How many replicates are ideal, 50-bp single end or 100-bp paired end, which one to follow, and the issue of coverage, how much coverage should be there. Now, coverage is calculated by a calculator, but this is valid for genome (DNA) sequences, transcriptome are different. Basically, we are not sure what parameters we should give to the sequencing facility to get even rare transcripts, including the ones in a gene family, get detected, in case they are differentially expressed. What should be the data size (million reads per sample)? Please advise. Thanks.
Hi Sach',
Coverage is not meaning full for RNAseq.
My advice :
- Illumina Paired-end stranded with Ribo depletion.
- 100 cycles ( reads length = 100nt)
- Should aim for 50 millions reads (x2 because it is paired-end) per library
- at least 1 biological replicate (meaning 2 samples for both WT and mutant).
Remarque :
for miRNA mutant you have to choose the tissues wisely to be sure that both miRNA and targets are expressed.
Be sure to apply the same culture conditions for all sample.
Alex
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Nucleic Acids
- RNA
- Small RNA
- microRNA