I am trying to analyze some environmental metagenomic samples. I would compare functional abundances between the samples for some genes of my interest, for example now, nitrogen cycle-related genes.
I initially planned to use MG-RAST server, but it is likely that it takes long time to finish the jobs on the server. It has been a month since I uploaded my sequence files, but the analysis is not finished yet. So I tried to perform blast for annotation by myself, but failed because of insufficient memory on my computer. I tried the program called "DIAMOND" and "MALT".
I am thinking that I could do it on my PC if I reduce the size of the database extremely by using custom database (i.e., creating fasta file only including nitrogen cycle-related protein sequences) instead of full NCBI nr database. Is this idea reasonable?
I initially planned to use MG-RAST server, but it is likely that it takes long time to finish the jobs on the server. It has been a month since I uploaded my sequence files, but the analysis is not finished yet. So I tried to perform blast for annotation by myself, but failed because of insufficient memory on my computer. I tried the program called "DIAMOND" and "MALT".
I am thinking that I could do it on my PC if I reduce the size of the database extremely by using custom database (i.e., creating fasta file only including nitrogen cycle-related protein sequences) instead of full NCBI nr database. Is this idea reasonable?
