Hello friends, maybe you can help me with this.
I have a coassembly of two metagenomes, from which I can retrieve the contigs that belong to a particular taxon of interest. My criteria for assigning a taxon to a contig is rather strict, asking for 80% of the genes in the contig to be undoubtedly assigned to the taxon. In this way I get >500 contigs summing up to 5 Mb, with some big contigs >100 Kb. Then I map the original reads to these contigs using Bowtie2, and extract the mapping reads. When I reassemble this set of reads using Spades, to my surprise I get considerably shorter contigs. Inspection with Mauve or nucmer shows extensive splitting of the original contigs, exactly the opposite of my expectations.
Could you please hint me on this? Could it be that the mapping step is producing an unbalanced coverage and as a consequence the contigs split?
Thanks a lot!
I have a coassembly of two metagenomes, from which I can retrieve the contigs that belong to a particular taxon of interest. My criteria for assigning a taxon to a contig is rather strict, asking for 80% of the genes in the contig to be undoubtedly assigned to the taxon. In this way I get >500 contigs summing up to 5 Mb, with some big contigs >100 Kb. Then I map the original reads to these contigs using Bowtie2, and extract the mapping reads. When I reassemble this set of reads using Spades, to my surprise I get considerably shorter contigs. Inspection with Mauve or nucmer shows extensive splitting of the original contigs, exactly the opposite of my expectations.
Could you please hint me on this? Could it be that the mapping step is producing an unbalanced coverage and as a consequence the contigs split?
Thanks a lot!