1
$\begingroup$

I have been trying to use a pipeline for predicting gene expression for a target gene (Although you can do it for many) - the name of this method is epiXcan. Here is a link to the paper:

https://www.nature.com/articles/s41467-019-11874-7#Sec22

One of the very first steps in the pipeline is obtaining SNP 'annotations' from an epigenome reference consortium known as the REMC. In the methods paper they say:

"We first prepare eQTL statistics (computed with MatrixEQTL4) and SNP annotations (extracted from REMC https://egg2.wustl.edu/roadmap/web_portal/). For each eQTL tissue, we use the matched REMC tissue to extract the corresponding annotations".

I have completed the first step of obtaining eQTL statistics. However obtaining SNP annotations is where I am lost.

In my case, I am only trying to predict the expression of one gene (DRD4) and so my region of interest is the pre-frontal cortex. If you visit the website they provide though, it doesn't seem as though annotating a large list of SNPs - at least in the way they are presenting - is possible.

Can anyone provide some insight or direction on how to proceed? How does one obtain SNP annotations from the REMC? I have tried emailing the authors of the paper and they have not been responsive.

Thank you,

$\endgroup$

1 Answer 1

0
$\begingroup$

After much research I found out how to obtain the annotations.

You must go to the REMC website: https://egg2.wustl.edu/roadmap/web_portal/

Go to the chromatin-state-learning tab, and choose the n-chromatin state model desired - the core has 15 states, then 18, then 25 (imputed states).

Following this, you must go to the download URL link and, you will find a lot of files, primarily with the following coding information: E0*.. This corresponds to a particular tissue/cell type and so you must choose the one relevant to your research question. The way they are correspond to each tissue/cell type can be found in the following paper:

https://www.nature.com/articles/s41467-019-11874-7#Sec22

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.