I want to go for whole genome sequencing using next gen Illumina but i have some problems.
Firstly, the organism is a mollicute with a small genome and is an obligate plant parasite with low titer. I cannot do a DNA extraction free from its host DNA. The best ratio i can possibly get is 1:10 bacterial to host DNA, optimistically. I can purify it using a NEBNext® Microbiome DNA Enrichment Kit but that will still leave me with organelle DNA.
The problem lies that the organism is very GC low, estimated around 26% GC, i've read that Illumina has a GC bias (Article Effects of GC Bias in Next-Generation-Sequencing Data on De ...
) ." For example, Illumina sequencing of a Plasmodium falciparum genome, which is extremely GC-poor with a mean GC content less than 25%, was found to favor the more GC-balanced regions, leading to few or no reads from the many GC-poor regions [13]. "
This is an issue because not only are we afraid Illumina will target GC regions of the bacteria but altogether favor the normal GC% host DNA which already outnumber the bacterial DNA. And thus we would be left with very low number of reads for our actual bacterial DNA.
Do you have any recommendations about this matter?
Also, in the event of not being able to extract adequate amounts of DNA, we would have to resort to an MDA, does anyone have experience with MDA favoring higher GC% genomes and also moving in favor of amplifying host DNA over bacterial DNA? . As a strategy to target my organism i can add oligomers in the MDA that are frequent in the organism reported in literature. Would that be recommended?
Would appreciate some answers.
Does it have to be Illumina? Maybe Iron torrent is better in this case? I suppose you have Illumina at your lab... Why don't you try and see what you come up with... It may very well work.
Pea aphid has a low GC content (about 35%) with a lot of bacteria symbionts and Illumina sequencing on DNA is working very well in this organism.
You'll just need to map your reads first to the host genome and map the unmapped reads to your mollicute genome.
You can do a test sequencing with really low number of reads, the proportion should not change much if you continue with deeper sequencing using the same library.
If you fear GC bias, depending on what type of method you use and what kind of quantification you want to make, I think you can correct for the GC bias.
This Link Maybe Useful:
Article Quantitation of next generation sequencing library preparati...
- Badges
- Science method