How to get coding sequences (CDS) of UniProt reference proteomes?

Question

I downloaded the following reference proteomes:

https://ftp.ebi.ac.uk/pub/databases/reference_proteomes/QfO/

'Drosophila':'UP000000803_7227.fasta',
'Arabidopsis':'UP000006548_3702.fasta',
'C.elegans':'UP000001940_6239.fasta',
'Zebrafish':'UP000000437_7955.fasta',
'human':'UP000005640_9606.fasta',
'mouse':'UP000000589_10090.fasta',
'yeast':'UP000002311_559292.fasta'

Now, I would like to have the corresponding coding sequences (CDS). My temporary solution is to download CDS files from ENSEMBL FTP server, translate those and match with the ref proteome. However, this only covers 50-90% of protein sequences.

Thanks!

Answer 1

I went through diverse README files and found the answer:

Coding DNA Sequence FASTA files (*_DNA.fasta)
=============================================

These files contain the coding DNA sequences (CDS) for the protein sequences
where it was possible.
The format is as in the following example (UP000005640_9606_DNA.fasta):

>sp|A0A183|ENSP00000411070
ATGTCACAGCAGAAGCAGCAATCTTGGAAGCCTCCAAATGTTCCCAAATGCTCCCCTCCC
CAAAGATCAAACCCCTGCCTAGCTCCCTACTCGACTCCTTGTGGTGCTCCCCATTCAGAA
GGTTGTCATTCCAGTTCCCAAAGGCCTGAGGTTCAGAAGCCTAGGAGGGCTCGTCAAAAG
CTGCGCTGCCTAAGTAGGGGCACAACCTACCACTGCAAAGAGGAAGAGTGTGAAGGCGAC
TGA

So, EBI itself provides the corresponding coding sequences, which they name _DNA.fasta. I misinterpreted the file-name as genomic (unspliced) DNA.