Jing-Zhe, It is known for several well-studied virus groups that a single gene can deternine its host range (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC230224/; http://www.nature.com/nature/journal/v270/n5638/abs/270617a0.html; http://www.ncbi.nlm.nih.gov/pubmed/8009842)

But, if your virus/phage is from new class you will not be able to figure out its host organism until you find it by experimental way. Phenotype goes first, genomics - second!  Sure, you can find some host -related sequences in the viral genome, but this method will work only if host species have known complete genome sequence.

Its genome can be identified by reading its host genome, because 99.99% of its genes are based on host genome. Integration of phages as well as virus in the nature is best way to find its sequence. A normal contig by contig sequencing with overlapping gene adds and relieve of from the main genomes by Pharmacia system sequencer or Sun system. 

As i see it, phages present something similar to dark matter. Very little is known about them.

So it is not surprising that phages even from same host have varied genomes and morphologies.

http://www.sciencedirect.com/science/article/pii/S0092867403002332

http://www.sciencedirect.com/science/article/pii/S0092867403002769

Thus just having a phage genome the chances are slim that you can identify the host unless you find some signature genes from its original host that might have moved to the phage genome by way of transduction.

Just curious how did you get the genome without its host ? ie how did you isolate the phage for genome sequencing ?

Jing-Zhe, It is known for several well-studied virus groups that a single gene can deternine its host range (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC230224/; http://www.nature.com/nature/journal/v270/n5638/abs/270617a0.html; http://www.ncbi.nlm.nih.gov/pubmed/8009842)

But, if your virus/phage is from new class you will not be able to figure out its host organism until you find it by experimental way. Phenotype goes first, genomics - second!  Sure, you can find some host -related sequences in the viral genome, but this method will work only if host species have known complete genome sequence.

Thanks for above answers!

To  Ravi,

I can't quit follow your answer. Do you mean I can sequence or analyze the host genome by Pharmacia system sequencer or Sun system? Would please provide some references?

To Jay and Alex,

Do you think most of the phage/virus's genome have similar sequences acquired from its host genome segments? 

I acquired the phage genome by only sequencing the virome (virus metagenome) isolated from a animal tissue, but I didn't sequece the whole metagenome.

Viral metagenome from host tissue? What tissue and what is the animal in this case.

If you are finding phage proteins then as I explained the original bacterial host might be very difficult to find. But if its a non-phage viron then using techniques like Alex suggested might help.

The only thing here is that IF you have made your metagenomic library from the faeces of the animal then depending on the animal being a herbivore or carnivore the virus can be from its diet as well.

some  phage/virus's genomes may have short  sequences acquired from its host genome, as well as bacterial host can have fragments of the phage in own genome. 

I would BLAST the genomic sequence into microbial database. Likely you will find homology to related phage or bacterial host sequence through which it integrates.