RNA-Seq in C. elegans

Model Organism: Worms Scientific Discussion
1

I am planning on an RNA-Seq experiment in C. elegans. One question that seems to be important for the sequencing facility is to know how big the transcriptome of C. elegans is.
in the literature, I found that approx. 70% of the genome is transcribed. Would that be correct for young adult animals?
Hope anyone can help me with this…
Thanks!

2

Which technology are you/the facility planning to use? I assume that your facility is concerned that the coverage will not be high enough. If you are using any of the Illumina NGS platforms (HiSeq or GAII) one lane per sample on a flowcell is more than enough. You could even barcode and run multiple samples on one flowcell, this depends on your specific question.
good luck.
Marlon

3

I was planning on using the SOLiD platform from Applied Biosystems. As for now we decided to load 6 samples per lane, with 15x single-end 50 nucleotide reads. I am mostly interested in expression levels/differential expression levels between two mutant strains during one stress condition. Is 15x coverage sufficient to detect also low expressed genes or would a microarray analysis almost be the better option in this case?
Thanks!

4

By 15X coverage, do you mean 30 x10e6 sequence tags per sample? If so, that should be sufficient to detect all but the least abundant transcripts (i.e., those with very low expression and/or expressed in a subset of cells).

If you intend to perform statistical analysis for differential expression (highly recommended), you’ll want to sequence biological triplicates. Sequencing would be more sensitive and (should be) less expensive than microarrays.

Harold

5

Hi Harold,
thanks for your reply. Yes, I will definitely go for biological triplicates. With 15x coverage I mean 15 million 50 nucleotide reads per sample - unless I misunderstood something…
Thanks,
Pat

6

Hi Pat,

Typically, X-fold coverage refers to the genome size, so 15X coverage for worms would be 15 x 10e8 bases. I divided that number by your read length (50) to calculate the number of reads (30 x 10e6).

X-fold coverage really doesn’t make much sense with RNA-Seq libraries (although people use it). For counting transcripts by RNA-Seq, the most important metric is the number of reads rather than the total number of bases.

Harold