[Ncbi Entrez] Retrieving Complete Genome Informations From Ncbi Genome

Question

[ls.duchemin]

Hello All,

I'm trying to build the correct Entrez query in order to get the informations for complete eukaryotic genomes from the NCBI Genome database. The genome browser (http://www.ncbi.nlm.nih.gov/genome/browse/) displays 185 entries when searching complete eukaryotic genomes.

I've been trying these :

• eukaryota[organism] AND complete[status] ; entries count = 319
• eukaryota[organism] AND complete[status] AND "genome sequencing"[Project Type] ; count = 300

Any ideas on either the best query to do what I want or which query corresponds to what is displayed in the browser ?

Thanks a lot !

[Leandro Lima]

Hello!

What kind of information do you want exactly?

Just the number of complete genomes?

[ls.duchemin]

No, I was trying to reproduce the genome browser output for complete eukaryotic genomes, using Entrez. That's why I started comparing the numbers of complete genomes, to see if my queries were corrects. Actually I want to get the informations like assembly ID, taxon ID, number of loci, % GC etc… for all complete eukaryotic genomes using BioPerl and Entrez. The problem is, if what I get through Entrez queries is different from genome browser's informations, which one do I choose ? And is there a query that would give the same output ?

Answer 1

[Neilfws]

I'm not convinced that the data on that page can be retrieved via Entrez.

If you follow the link to the FTP site and download the file eukaryotes.txt, you'll see a field named Status. This is where the value of 185 comes from - I opened this file in R:

euk <- read.table("eukaryotes.txt", header = T, sep = "\t", stringsAsFactors = F, comment.char = "", quote = "")
table(euk$Status)

#         Chromosomes              No data Scaffolds or contigs 
#                 185                 1609                  722 
#       SRA or Traces 
#                 455

However, if you experiment with the Advanced query builder at the NCBI website, you'll find that:

• database Genome has field Status, but "chromosomes" is not a valid value
• databases Bioproject and Assembly do not have field Status

So it may be that there is no direct relation to the Entrez databases. Or I may be wrong and it's just very difficult to formulate the query :)

[ls.duchemin]

That's right but it feels weird that NCBI doesn't use the content of its databases to generate this file... I started using that file, since it already contains most of the informations I need. It's just, I'm not very comfortable with working on it while not knowing how its generated and if it corresponds or not to NCBI databases content.

Edit : An interesting fact is that "eukaryota"[organism] gives me like 2100 lines and the eukaryotes section in genome browser is more like 2900…

Answer 2

[User]

The post was deleted.

[Neilfws]

But that gives 32 results; we're looking for 185. And you did not specify eukaryota.

[skymningen]

Then try eukaryota[organism] AND complete[status] AND "has chromosome"[properties]?

[Neilfws]

That gives 128. Is no-one trying before posting :)