Hello experts,
I am trying to BLAST my Illumina reads (paired-end reads with a length of 100-bp) to antibiotic resistance gene database. I have seen some of papers doing this but all of them used raw reads from Illumina or 454 rather than assembled contigs for BLAST. So I am wondering if it is better to use raw reads for gene search (ARG in my case) rather than contigs to avoid assembly bias? Or is it better to use contigs which are longer than raw reads thus increasing the length of alignment length? Any advice would be greatly appreciated. It would be also great if you have seen any paper that uses contigs for this purpose.
I am trying to BLAST my Illumina reads (paired-end reads with a length of 100-bp) to antibiotic resistance gene database. I have seen some of papers doing this but all of them used raw reads from Illumina or 454 rather than assembled contigs for BLAST. So I am wondering if it is better to use raw reads for gene search (ARG in my case) rather than contigs to avoid assembly bias? Or is it better to use contigs which are longer than raw reads thus increasing the length of alignment length? Any advice would be greatly appreciated. It would be also great if you have seen any paper that uses contigs for this purpose.