Quality of Assembly based on Ion Torrent Data (microbial genome)

[Biswapriya]

Hi All !

After some good exercise with data, I obtained following results as follows: Ion-Torrent Data for a 5 Mbp microbial genome: Acquired 3, 435, 101 high-quality filtered reads, average read length of 180 bp, assembled using MIRA/ CLC Assembler, & got: 350 Contigs (> 500 bp CO), mean contig size: 17 kbp, 110 X Coverage; N50 Contigs: 32 kbp, largest scaffold: 89 kbp, 6000 ORFs/CDSs.

What looks unreasonable/ unacceptable/ gross ? Please feel free to advice and comment ! Just wish to check out and find flaws.

Thanks in advance,
Biswa

[jonathanjacobs]

For single read 200bp data - In our experience with the PGM running 200bp reads - that's about right. You'll be lucky if you break 100K for your largest contig for most bacterial genomes with single end reads on the PGM.

Your best bet would be to jump to 400bp reads. The larger chips won't really help much either, so stay with 316 or even a 314 and save some cash.

Alternatively - you can go for mate-pair sequencing; which on the PGM is actually quite good with the right library prep.