Taxonomic Annotation from orfs

[tpellegrinetti]

Hi all, Sorry if this question is silly.

I have several fasta files from some kegg orthogoly of my interest which I extracted from prodigal and hmmsearch result. For example, the nifH, that is the K02586.faa.

This file started as follow: ">L04SR_181+k119_40695_8 # 7004 # 8506 # 1 # ID=40466_8;partial=00;start_type=ATG;rbs_motif=GGAG/GAGG;rbs_spacer=5-10bp;gc_cont=0.457"

I would like to make a custom database from this fasta orf files and then make the taxonomic classification in metagenomic reads using Kaiju.

It's possible?

I tried to do this but it not works for me. Follow the script that I used:

kaiju-mkbwt -n 5 -a protein -o K16395 K16395.faa kaiju-mkfmi K16395 kaiju -t /Storage/databases/KaijuIndexes/nodes.dmp -f K16395.fmi -i /Storage/data1/thierry.pellegrinetti/05_metagenomas/2018/L04SR18_1.fasta -o L04SR18_1.out kaiju2table -t /Storage/databases/KaijuIndexes/nodes.dmp -n /Storage/databases/KaijuIndexes/names.dmp -r genus -o L04SR18_1_summary.tsv L04SR18_1.out

[pmenzel]

When you want to make a custom database for kaiju, you need to follow the format described here: https://github.com/bioinformatics-centre/kaiju#custom-database The fasta-headers need to contain NCBI taxonomy IDs, that are also present in nodes.dmp and names.dmp from the NCBI taxdump files.