notes on the collaborative manuscript on GWAS risk studies

Here are a few notes about contributing to the Nature Genetics manuscript that I was talking about in a previous post.

note2: I have opened a discussion on Biostar, if you are interested in contributing, look there also.

Scope and purpose of the article

The main purpose of the article is to explain how the results from a GWAS study can be functionally validated. Let’s say that a study has identified a SNP variant that is likely to be associated with a trait: the collaborative article describes the methods that can be used to demonstrate the association, by identifying eQTLs and study them through microarrays to building animal models to simulate the effect of the variant.

In my opinion the key to understand what the manuscript is about, and why it is being written collaboratively, lays in this recent Nature Genetics editorial:

the authors of the Editorial say that most of the times, after a GWAS study has found association between a SNP variant and a risk for a trait, the result is not followed by a functional characterization of the SNP.

(edit) Moreover, you should also look at the home page of the group that has written the original draft, which is the Post Genome Wide Association Study Initiative.

It seems that the same Nature Genetics authors are sponsoring this article as a way to promote discussion about the future after GWAS studies. Instead of proposing to some selected authors to write a review on the topic, they are calling for help from all the scientists interested on Internet. I think it is a nice idea to promote discussion.

Ideas for new paragraphs

I thought of at least two paragraph that could be introduced in the article. I talked about them in the Discussion page of WikiGenes.

The first is to talk about the computational approaches that can be used to predict the function in which a SNP variant is involved. A gene prediction tool can determine whether the variant is within a non-annotated gene or pseudogene. For variants falling in a coding region, there are many approaches to predict the effect on the protein structure or modifications.

The second idea is more related to what I do in my lab, and it is about making use of the information on pathway and gene-gene interactions to improve the results of the study. The goal is that if we are able to understand how selective forces distribute over a pathway of genes, we can make better predictions on how a variant in a certain gene can affect the phenotype if we take into consideration the role of that gene in the pathway. It is kind of a better way to calculate the background distribution, to understand how much variability I should expect from a gene knowing its function. I am writing about this and putting some references in the discussion page and then in the article.

6 Comments

  1. Hi,

    Anita Göndör and undersigned added a paragraph on chromosomal networks and how these may propagate epigenetic effects in a genotype-specific manner in the “epigenetics” section of the manuscript. Given the conundrum that 88 % of GWAS SNPs map in gene deserts/intergenic regions I do think that the perspective of chromosome interactomes is needed in such an important contribution. If this addition is welcomed by the original authors and if there is space we can elaborate on this topic.

    Please let me know what you think.

    best wishes,

    Rolf

  2. Hello Rolf!!
    I saw the paragraph you put in the wiki. I think it is relevant to the manuscript, especially for what you say, that 88% of the GAWS SNPs map in non-genic regions, and because there it was already a paragraph on epigenetic effects.
    It is unfortunate that the original authors of the manuscript are not answering too much to the doubts in the Discussion page.. if they were more active, it would be easier to make good contributions (leading to a better article).

  3. Hi,

    Thanks. I have myself been quite unsuccessful to make GWAS specialists understand the importance of chromosome crosstalk in the context of GWAS. I guess that the line of thinking in three dimensions is currently not trendy in the GWAS research area. This may probably have to change if the aim is to understand the function of the loci identified in GWAS screens.

    There are so many different issues emerging from the 3D perspective of GWAS. One is that chromosome interactomes are poorly evolutionarily conserved making it less meaningful to use animal model systems for human diseases. Another is the functional commonality of the interactome nodes that might be candidate loci in GWAS screens. ETC.

    The problem is that Anita and I saw this opportunity to spread the gospel of chromosome interactomes in GWAS a wee late and do not really have the time to write too much on this manuscript from next week and onwards. As I have not received any response from the original authors it is quite possible that the paragraph and the angle provided by the paragraph that Anita Göndör and I wrote will have to be deleted in the final version of the manuscript. That would be a pity though.

    best,

    Rolf

  4. ok, I can try to ‘support’ your contribution, even if there is no direct way to talk with the authors. I can try to integrate it better with the rest of the manuscript: for example, what you said about the fact that poor conservation of chromosome interactions makes animal models less useful can also be put in the “Models for testing function” chapter of the text. I’ll look at your papers then.

    cheers,
    Giovanni

  5. Hi!

    What’s the state of the art right now??
    If I understood correctly, the article is open only for discussion about the completion of the paper.
    Thanks in advance,
    Take care

    Yari

Leave a Reply to Yari Ciribilli Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code class="" title="" data-url=""> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong> <pre class="" title="" data-url=""> <span class="" title="" data-url="">