Hard to automatically download NCBI protein IDs

Question

[The]

Hello,

I've a few NCBI protein ids which are really hard to download with BatchEntrez or NCBI "datasets" for example

MBN1612259.1 MBF9061340.1

Though the seem to exist in good shape:

https://www.ncbi.nlm.nih.gov/protein/MBN1612259.1 https://www.ncbi.nlm.nih.gov/protein/MBF9061340.1

However when trying to retrieve by BatchEntrez , getting the following error:

Error Message

Can anyone tell me how to download such sequences automatically?

Thanks in advance

Answer 1

[Mensur Dlakic]

These commands work:

efetch -db protein -id MBN1612259.1 -format fasta
efetch -db protein -id MBF9061340.1 -format fasta

>MBN1612259.1 MAG: carbamoyl-phosphate synthase large subunit [Polyangiaceae bacterium]
MPRRDDIQKILLIGSGPIVIGQACEFDYSGTQGAKALVGLGYDVVLVNSNPATIMTDPELVRRTYVEPLE
VNTLAAIVARERPDALLPTLGGQTALNLALELHESGVLSQHGVQLIGAQVEAIRKAEDRQLFKDAMAHAG
LECPKSGYARSGEEARDIAVLTGYPLILRPSFTLGGAGGSVVDGPEQLEERVQWALAQSPTREVLVEESV
IGWKEFELEVMRDRADNFIVVCSIENIDPMGVHTGDSITVAPAMTLTDREYQCLRDASRAVMHEIGVETG
GSNVQFAVDPKTGRVLVIEMNPRVSRSSALASKATGYPIAKVAAKLAVGFTLDELENDITGTSAAFEPTI
DYIVVKWPRFAFEKFPGSDPRLGTQMKSVGEVMSIARTFPQALQKAARSLETGKDGLTSLFGRIDYVSMA
AQRTDKRDLALEPPPAARPRPSRAPGLRDEMARALRAIVGTPTAERLFHVADAIRLGVSIDELAQLTGID
SWFLGQIDRIVQHERVLREAPELDRVLLWESKRLGFSDRQIARLRDTDEAAVRLQREQAGIGTVYQRVDT
CAAEFVARTPYLYSSYETDTESEVSERRKVIILGGGPNRIGQGIEFDYCCCHAVFALRELGYETVMVNCN
PETVSTDYDTSDRLYFEPLTLEDVLAVCKEEASRGELVGVIVQFGGQTPLKLAVPLERAGVRLLGTSADA
IDRAEDRERFDALLNKLGLLRPRAGIAKSLDEARWVVGDIGYPVLVRPSYVLGGRAMMICWSDEELDAYV
GLALEAAQDEDSSPTLLIDEFLKHAIEVDVDCIADGHRAVVGGVMQHIEEAGIHSGDSTCVLPPHSLEPS
VVAEIEAQAKALALELGVVGLMNTQFAVKDGRVYVLEVNPRASRTVPFVSKATGRPLAKIAAQLMVGKTL
DELELRDLPQPTHVAVKESVLPFAKFPGVDTILGPEMRSTGEVMGVATSMPLAFSKSLLSAGSKLPARGR
AFISVTDEDKPAACYVAMHLRNVGFTVVATDGTADALTRARIPAVRINKVRQGSPHIVDAIKSGSIQLVI
NTAREASAIRDSYAIRRHAVLGNIPYFTTMSAALAVVESLEAHTLLDSRSVPVRSVQEWHARARRAARPE
>MBF9061340.1 Na(+)-translocating NADH-quinone reductase subunit A, partial [Rhodobacterales bacterium HKCCSP123]
ERPAAILVMGCDTRPLAPHPAEALAGREEALSRGLQALTGLTEGPVFLCDDAVRPLGVTVPGVRIVATVA
RHPQGLAGIRIAALCPAEIDHPVWDLDAEDVADLGDLLATGHLPQRRVVRVAGPALTETRLVTCQIGADT
RGLSYGAIRPGPHVILAGSPVDGRPAHWLGPRDRQVTVMDAAPRAAAPHWFLAALTRSSRPRPLIPSAAV
MQAAGGAFPAMAMLRALGAGDDETALKLGALSLLEEDLALVDYVTGGRPRAAELLRALLDRTAAEAGQ

[The]

Thank you very much. For Efetch , is there a way I can pass a file with multiple IDs as input ?

[GenoMax]

You could use epost method for multiple inputs (in this case it does not appear to work for some reason with MAG ID's you have). So write a for loop to pass the ID's to efetch one at a time.

[The]

is there any limit to download by efetch in one particular session?

[GenoMax]

How many ID's do you have? As long as the number is not in thousands you should be fine. But if you have a ton of them then consider using an alternate method such as extracting them from nr database.

[Mensur Dlakic]

If you make a file called lst that contains all protein IDs, this command will save all FASTA sequences in sequences.fasta.

cat lst | xargs -i efetch -db protein -id {} -format fasta >> sequences.fasta

Answer 2

[josev.die]

If you are familiar with R :

#Dependencies
library(refseqR)

# Get the protein sequence using multiple ids. 
accession = c("MBN1612259.1", "MBF9061340.1")
my_aa <- refseq_AAseq(accession)

# Make a fasta file 
Biostrings::writeXStringSet(x= my_aa, filepath = "mypath/aa_result")

Then, you can just read the file .