How To Get The 3' Utr Given Geneid And Transcriptid In Ensembl Using Python?

Question

[Sam]

Hello Everyone,

I have to do data collection for my current project where i have to collect all the 3' UTR sequences given Ensembl GeneID and Ensembl Transcript ID. I am new to this field and was wondering whether there is an easier way to do this than manually getting each 3' UTR. I have a list of GeneID and Transcript ID.

I did explore Ensembl BioMart, however i could not figure out how to exactly input both GeneID and transcriptID.

Also, would there be a way to incorporate this into python? I do know SQL at Beginner-Intermediate level.

Thank you very much in advance.

Answer 1

[alex.rubinsteyn]

A possibly more convenient solution: https://github.com/hammerlab/pyensembl

Answer 2

[Darked89]

Biomart: http://www.ensembl.org/biomart/martview

Database: Ensembl Genes 61 Dataset: Homo sapiens genes

Filters: check box on ID list limit There is a text area where you can paste Ensembl gene ids, i.e.: ENSG00000139618

Attributes: Sequences

Then below select: 3' UTR

Click count (top left of the page) just for checking, then go for Results.

Answer 3

[Michael]

You do not need to input both GeneID and transcriptID at the same time, the transcript id should be sufficient and it is unique. It is true that a gene can have multiple transcripts, but a transcript should only be assigned to one gene, so with a transcript id, gene ids are redundant.

Biomart has some built in tools to automatize querys, here is an exported URL to a Biomart query, such as the one described by darked, but with transcript ids as a filter instead of Ensembl gene ids. If you follow the link, you have a query you can start with. You can also get this query as XML or directly as a perl script. If you really must use python, there is more documentation on how to use the REST or SOAP interface.

Answer 4

[Sam]

Darked89 and Michael Dondrup: thank you for answering my question

Michael: I tried it the way u suggested. However i only get back 1 result after inputting all my transcript ids. example link with multiple ids a bit shorter

Here is the [file with all the TransID][2]

Thank you for your time.

[2]: http://www.filedropper.com/transid "file with all the transID

[User 59]

Argh! I realise your reply was too large to paste in as a comment, but please do not add an answer if you're trying to address comments. If you want to paste a large amount of text into a comment on someone else's reply, please use something like http://pastebin.com/ and point a link to the output you would like us to look at!

[Michael]

No, your link works fine, the result is a multi-fasta file, scroll down ;)

[Michael]

He tried to do it right, but the very long URL broke the formatting. I put a shorter URL.

[Sam]

Daniel Swan: I'm sorry about making a new comment. I am new to this website and i did not want to paste the huge link to the comment section. Therefore i thought i would make a new comment. I will keep this in mind in future. Thank you

[Sam]

Michael: Thank you very very much. I realized after looking at your shortened link that i was not putting commas after every transcript ID. Therefore it was not querying all the ID's. You put in 5 ID's separated by commas and therefore it queried and returned all of them. I put all my ID's in that format and i got the answer! Thank you very much for all your time and effort. :)