Entrez esearch.fcgi large set of sequences download: fluctuating number of sequences

[Toliman]

Hello everybody!

I have a little problem: I'm trying to download a large set of fasta sequences from Entrez, but the number of sequences retrieved differs from the announced number:
When I am using the link in a browser:

https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?db=nucleotide&term=Phaeophyceae[Organism]&usehistory=y

It will give me a certain number of sequences.
But when I use the perl script (as explain here: https://www.ncbi.nlm.nih.gov/books/N..._esayers-5-4-3), the number of final fasta sequences is always lower...
I am using just a small difference compared with the tutorial:

        $efetch_url = $base ."efetch.fcgi?db=nucleotide&WebEnv=$web";
        $efetch_url .= "&query_key=$key&retstart=$retstart";
        $efetch_url .= "&retmax=$retmax&rettype=fasta&retmode=text";

Does anybody had the same problem before? Does anybody know where does it come from and how to fix it?

Thanks in advance,
Denis

[Toliman]

Ok, I found the problem: DO NOT USE eUtils !

Use Entrez Direct instead, It's really working (be sure to install the latest NCBI version and not the aptitude package)
https://www.ncbi.nlm.nih.gov/books/NBK179288/

[GenoMax]

NCBI requires use of API tokens with eUtils as May 1st, 2018. It you are running a large amount of queries some of those may fail, which may explain the variable numbers. Just a thought.