I am doing an experiment to check the differential expression of bacterial genes under different drug treatments. I have 2 replicate.o
samples. I have two questions:
1. Which platform should be best- Miseq or whole lot data from Hiseq
2. Which aligner should be best- Bowtie or BWA
Any suggestion or personal experience working on such experiments will be very helpful
Thanks
If my understanding is correct , you are looking forward for a whole genome sequencing for your samples(or may be you are looking for your gene amplicon's sequencing alone), for which i would say you can go with a platform that can provide you a lot of information . So very likely Hiseq2000 can give a huge volume of data for all your samples at a very high coverage depth.This will be very useful in studying change in expression of group of genes without missing any.
This might be a out of way suggestion, if you are looking for longer read lengths then you can move away from Illumina and try Roche 454 GS-FLX + which can give you about 700bp long reads, but the throughput is relatively less.
Having said all these , your budget plays a role as well...
Yes you are right I am doing whole genome sequencing. I agree that budget is imp. However since bacterial genome is small do we need Hi seq data? I was debating Hi seq may be just an overkill and I can get what I need reasonably well withy Miseq.
Thanks
Miseq would give a good throughput as well. More data more satisfaction..still Miseq is not a bad option either since your bacterial genome is smaller.. But if you ask me, I would say Hiseq has a slight edge over Miseq..
After tallying up the number of replicates and controls needed multiplied by the sequencing depth required for transcriptome sequencing the required number of reads needed off of a machine can add up really quickly for a proper experiment. If I recall correctly I think somewhere in the neighborhood of 2-5 million reads per sample (assuming a 2-4 mb genome) per replicate is a safe number to shoot for. Thus multiple MiSeq runs. All this also depends on how efficient your rRNA depletion works on your particular species.
If you have equal access to both, Miseq versus Hiseq is really a money question, not a scientific one. But you probably won't need more than 10 Mreads/sample, so you won't need more than a fraction of a Hiseq lane for each sample.
bowtie and bwa are going to work pretty much the same.
Longer read length isn't going to help you. You'd much rather have more short reads than many long ones. They only need to be long enough to let you align uniquely, and you don't need 150 bases for that.
Yes you are right I am doing whole genome sequencing. I agree that budget is imp. However since bacterial genome is small do we need Hi seq data? I was debating Hi seq may be just an overkill and I can get what I need reasonably well withy Miseq. Thanks
Miseq would give a good throughput as well. More data more satisfaction..still Miseq is not a bad option either since your bacterial genome is smaller.. But if you ask me, I would say Hiseq has a slight edge over Miseq..