blast metagenome

[morning latte]

Sorry that this is a very simple question that might be already posted but I could not find the answers.

I have bacterial metagenomic sequences generated from Illumina HiSeq.

I am done with assembly of short reads and am about to blast them. Since I am new to this field, I am not very clear with the next step.

Q1. For blast, I first used the command below.
formatdb -i input.fasta -p F -o T

It generated several files with different extensions.

Then,
blastall -p blastx -i input.fasta -d my_database -o my_output_file

I guess "-i" is the file I used for formatdb command. What should I use for "-d" as I have several files generated from formatdb command?

Q2. I found that someone uses prokka before blast. I haven't heard about this pipeline but I assume it annotates sequences. Should I use this kind of annotation pipeline before blast?

Thanks for your help in advance!

[GenoMax]

I am not sure what exactly you are trying to do but you can look at PAUDA which is meant to be used for metagenomic protein database search tool. http://ab.inf.uni-tuebingen.de/software/pauda/

As for the blast it is useful to peruse the command line manual to find answers for some of your questions: http://www.ncbi.nlm.nih.gov/books/NBK1763/ BTW: You appear to be using an older version of BLAST suite (current program for creating blast indexes is makeblastdb).

Blast indexes comprise of multiple files. You need to only use the prefix of the files (generally the -o name that you used when creating the indexes) while running the blast search.