Anyone performed 454 sequencing on Amoeba?

[cheerioet]

Hi,

I had met an investigator, whom would like me to propose sequencing on amoeba. I am quite new with such research project, and would like to ask your expertise advise regarding on how to perform sequencing for amoeba.

1. The investigator had some isolates, and would like to perform NGS to find out if those isolates are identical or similar to reference genome?

2. The investigator would like to know, if their objective is to run metagenomic for amoeba diversity and population is certain niche area, how can it be done?

I sincerely appreciate everyone's help and suggestion.
I desperately needed some information as I really have no idea of how to start this.

Thanks in advance.

[rhinoceros]

I'd probably go for amplicon sequencing. First things first, you should really do some literature research. It might also be worth considering if your expertise is up to the job..

[SNPsaurus]

There are a few choices. Some amoebas have huge genomes (700 Gbases!) while others are small. Very large genomes (>10 Gb) will rule out all but amplicon sequencing. For moderate genomes, some kind of "less than whole genome" approach would make sense.

Genotyping by sequencing was invented for this purpose, to sample 10-100k loci across a genome in a reliable way. It is a bit of an acronym soup, with RAD-Seq (from my academic lab), GbS, ddRAD, REST-Seq and nextRAD (commercial bias warning! from SNPsaurus). For these approaches, each isolate would have >10k tags sequenced to a moderate depth, and the genetic variation between samples and to the reference identified. Pools of samples can also be sequenced and the allele frequencies measured as a metric of population diversity.

People also do "skim sequencing" which is making a whole genome shotgun library and then sequencing to less than 1X, but this makes the data analysis more difficult, and since that is usually the hardest part anyway better to not make life even more difficult!

For amplicon sequencing, you would want to assay several regions of the genome by PCR, and hope that those regions are typical for polymorphism frequency, then do the same kind of analysis as genotyping by sequencing but on a more limited data set.

[krobison]

Knowing the genome size within an order of magnitude is essential. Also, why use 454? You will pay a much higher price per read than with other technologies. Perhaps for a long amplicon it will be better (though MiSeq & PacBio CCS are encroaching rapidly), but for any sort of shotgun approach you are much better off with Illumina (if the genome is mid to large) or PacBio (for small genomes)

[cheerioet]

I appreciates the comments and suggestion.

RAD seq like in http://www.nature.com/nrg/journal/v1...df/nrg3012.pdf ??

Investigator would like to have longer reads, in which could reduce the complexity of bioinformatic later on.

If I do not mistaken, the studied subject genome is about 45Mb.

[chadn737]

At 45Mb just sequence the entire genome with Illumina. Depending on the number of samples, you could easily do this at low coverage or even high coverage. 454 is not going to make the bioinformatics any easier.

[cheerioet]

hmm... indeed, i ain't bioinformatician which i felt i will have difficulties handling the data...