I’m new for using DIAMOND and MEGAN. I already run the alignment for my samples using DIAMOND, but the result is reducing almost 25%. I have only get 75-80% queries aligned with DIAMOND? Is it still good to perform it in MEGAN?
I also get a .daa file from the alignment, but i wonder how much file i should get when i perform 4 samples? In my case, i only get one .daa file from 4 samples.
Here is command that i used for perform the alignment:
diamond blastx -d nr -q HA_result/final.contigs.fa -o matches.daa
diamond blastx -d nr -q HB_result/final.contigs.fa -o matches.daa
diamond blastx -d nr -q IA_result/final.contigs.fa -o matches.daa
diamond blastx -d nr -q IB_result/final.contigs.fa -o matches.daa
don’t give the output for different runs the same name. Here you are using the name matches.daa for all four runs, so the next run overwrites the output of the previous run…
You need to use the format option to ensure that the output is indeed written in DAA format, so use this:
-f 100
In summary, you should be running something like this:
diamond blastx -d nr -q HA_result/final.contigs.fa -o HA_result.daa -f 100
diamond blastx -d nr -q HB_result/final.contigs.fa -o HB_result.daa -f 100
diamond blastx -d nr -q IA_result/final.contigs.fa -o IA_result.daa -f 100
diamond blastx -d nr -q IB_result/final.contigs.fa -o IB_result.daa -f 100
Also, it looks like you are aligning contigs, not reads, so you should use DIAMOND’s long read mode, see the latest DIAMOND documentation to determine the exact option.
Thank you so much for reply. It’s really help me and other who find the same case. Yes, I use contigs for aligning. Oke thank you Daniel, I’ll read for long reads. Thanks for your advice.