What I did :
1. I have manage to concentrate the targeted bacteria (doesn’t have reference genome available) in the suspension before DNA extraction. I made 16s rRNA clone library which suggest that the targeted bacterial population is almost 50% of the total population.
2. Sequenced the sample by Illumina MiSeq genomic shotgun paired end read (2 x 300bp) sequencing.
3. I use MetaSPAdes to generate assembly and then MetaBAT for binning.
My main Question here is that, what post binning strategy I should follow to get the full/partial genome of the targeted bacteria.
Each bin is equivalent, roughly speaking, to a genome.
You should use CheckM to inspect the lineage of each bin, GC content, the degree of contamination and completeness of the genomes. FinishM can help you to close gaps between contigs of a single bin so you get a less fragmented genome
May be this will help
https://microbiomejournal.biomedcentral.com/articles/10.1186/s40168-016-0154-5
from my experience, if you want to get a full/whole circular genome based on your bin. you have to do manual finishing usingthe linkage information among contigs in your bins based on paired-end or better mate-pair reads mapping. you could read more from here: http://madsalbertsen.github.io/multi-metagenome/docs/step12.html
If there is microdiversity in your bin that cause the assembly to be very fragmented, say >200 contigs in a bin. Then whole genome could be hardly get unless you have large insert linbrary (mate-pair library) or some long sequence (pacbio or nanopore) to assistant your finishing. Alternatively you should try to improve your initiral assembly by reduing the affact of microbiversity, the most easy way to do so, is to reduce the size of data used for initial assembly.
The short answer is that you can't get a full bacterial genome using short reads. The reason is that the repetitive regions larger than your read length won't be defined. Now, if you have 2x300 reads, what is your insert size? if the reads are overlapping (say 550 insert size) you can use DiscovarDeNovo to assemble it and probably you will get a very decent assembly of 50 contigs. However, for most of the finishing tools available, you will need non-overlapping reads to scaffold and bridge the contigs.
As mentioned, if there is diversity, that will fragment the genome. If the other bacteria that are not your genome of interest, have a reference genome, map your reads to them and select those that doesn't map or those with not so good mapping. Probably those are reads that will belong to your genome.
Good luck.
Thanks Xabier, Yu and Alejandro for you suggestions. At the moment I am trying to inspect my bins (generated by MetaBAT) and see what we got.
- Badges
- Science method