Pulling data from NCBI Nucleotide database

[yaaminiv]

I want transcriptome FASTA sequences from the NCBI nucleotide database (BioProject PRJNA74409, NCBI TSA accession numbers JR970414–JR999999 and JT000001–JT023377). Is there a programmatic way to grab all of these sequences at once? I found this link that suggests using eutils, but I'd have to create a comma-separated list of 52963 accession numbers, and the NCBI Eutils page suggests using "the HTML POST method" if searching for more than 200 accessions (I also have no idea what they mean by HTML POST method). Any suggestions/insights on how to get these sequences easily?

[kubu4]

This notebook post might be helpful (it's very similar, but not exactly what you're doing):

https://robertslab.github.io/sams-notebook/2022/07/06/SRA-Data-S.namaycush-SRA-BioProject-PRJNA674328-Download-and-QC.html

Look it over and see if you have more questions and I'll be happy to help!

[kubu4]

Actually, you can ignore my previous post. Do the following:

1. Go to BioProject page.
2. Click on the "Number of Links" number of transcripts in the Project Data table.

Then, send all of the info to FastA format:

[kubu4]

That'll get you this FastA file:

head sequence.fasta 
>JT023377.1 TSA: Acropora millepora Cluster028086.Acmimixed mRNA sequence
AAGGTTCCAGAGCTTAATTCTATTGCGTTAGAAATTTTCCAGAAATGTATGTTAAATGGGATCACTATTG
ATGTGAATTGGATCCCAAGAGATTTCAACAGTGTGGCAGATGAGATTAGTAAGATAATAGATTACGATGA
TTACACAATTAATGATGATATTTTTGCTTTTTTGGACAAATCATGGGGACCCCATACAGTCGATCGTTTT
GCGTGTCACTATAATAAGAAGCTACCTTTATTTAATTCGAAGTTTTTTCAGCCAGGCACGAGTGGGGTGA
ATGCTTTCAGCCAAGACTGGGCTTTTGCCAATAATTGGTTATGTCCTCCCATTTATCTCACCGCGAGGGT
AGTTAATCATTTGAAAGCTTGTAGAGCGGCTGGGACCCTTATCGTCCCTTTGTGGAGATCGGCACATTTT
TGGCCAATTATTTGTGACGATGGAGTTCACTTTAGTAACTTTGTGCATGACTGGTATGTTTTGCCGCACA
TTCCTAATTTATTTATTAGAGGTAAAGCAAAGAATTGTATCTTCGGAAACGGCCAGTTGAAGTTTATAAT
GTTGGCATTAAGGATTGATTTTTCTGTCCCTCTTAGGTCCGCTGTACGAGGATTTTGTACTGAGTTTAAG

Count all the entries in the FastA file (i.e. all lines beginning with >) to make sure we got them all:

wc -l sequence.fasta
52963

[yaaminiv]

@kubu4 Thank you! If only the NCBI website wasn't a dumpster fire I probably would have figured it out myself 😂