Contribute To The Nat Gen Paper "Principles For The Post-Gwas Functional Characterisation Of Risk Loci"
6
8
Entering edit mode
14.0 years ago

The Post Genome-Wide Association Initiative has recently launched a nice initiative: they are writing a perspective on how to functionally characterize the variants identified by a GWAS, and they made the draft of their paper publicly editable, so anybody interested is able to contribute or review it before the publication.

Over the last years, many GWAS have identified variants involved in cancer or other congenital diseases; however, too frequently these GWAS are not followed by a proper analysis to characterize the mechanism that relate the variant with the disease (the etiology), or to identify the real causal variant.

So, the paper will be a perspective to describe methods and best practices to characterize the functions of a SNP or variant identified by a GWAS.

The paper is well written and is already most complete, but I believe it lacks a chapter on the characterization of the function of a variant using computational tools. For example, a gene prediction tool can be used to predict whether the variant identified lays into a non-annotated gene or pseudogene, and other software can predict whether it is in a regulatory position; or, in the case that a variant falls into a coding sequence, there are many tools to determine its effect on the structure of the protein.

So, you can find an editable version of the draft into the WikiGenes page. However, I propose you to do the following: write here, in Biostar, about any computational method that you know of or that you could apply, so we will be able to vote the best contributions and discuss them here, avoiding confusion in the wiki. The deadline for the manuscript is December 20th, so we will choose the best answers a week before that, and put the in the draft. The authorship will be recognized, and the posts on biostar will be always stay here to testify your contribution. Edit: I meant that your authorship on the wiki will be recognized, and we will tell the original authors of the draft to look at this question in biostar. However, whether to include you or anybody as authors of the paper is a decision that depends from the corresponding author.

p.s. I am also putting down some notes on my blog. If you want to contribute to the paper, look also at the Discussion page.

gwas snp • 9.0k views
ADD COMMENT
2
Entering edit mode

"a GWAS study" is the same as HIV virus or LCD display ;)

ADD REPLY
1
Entering edit mode

I want to share some of my recent experiences in post-GWAS analysis and would love to share it here for comments from BioStar members. But it is not clear to me how collaborative contribution is possible in this particular case ? I think the Wiki will only recognize individual contribution by distinct username. Is there a way to inform the coordinators that the content is a group effort ?

ADD REPLY
0
Entering edit mode

@Khader: the best way is to talk to the corresponding authors and tell him to look here.

ADD REPLY
0
Entering edit mode

I will write to the authors and check this.

ADD REPLY
0
Entering edit mode

ok!! remember that you can also contribute to the document at WikiGenes directly. I was just proposing to create a discussion on computational methods here at Biostar, to involve the Biostar community, but if you have something that could be submitted into the original document directly you should better go to WikiGenes and edit there.

ADD REPLY
0
Entering edit mode

ARRGGHHH!! We bumped into each other editing over there. I really liked the idea of editing here and then bringing it over...alas...

ADD REPLY
0
Entering edit mode

don't worry, I reintegrated my previous changes... did you loose something in the process?

ADD REPLY
0
Entering edit mode

No, I'm fine....I am just easily frustrated by wikis...

ADD REPLY
0
Entering edit mode

I am not able to submit the sections to the article I am getting "Text exceeds maximum length The text you are attempting to submit is too long. Please shorten and save it again." error - any idea why this is happening or is there any limit on inserting text ?

ADD REPLY
4
Entering edit mode
14.0 years ago
Mary 11k

Oh, yay! I put this on our blog too but wasn't sure how to put it in the form of a question over here :)

GRAIL: blog post about it here; citation is: Raychaudhuri, S., Plenge, R., Rossin, E., Ng, A., International Schizophrenia Consortium, Purcell, S., Sklar, P., Scolnick, E., Xavier, R., Altshuler, D., & Daly, M. (2009). Identifying Relationships among Genomic Disease Regions: Predicting Genes at Pathogenic SNP Associations and Rare Deletions PLoS Genetics, 5 (6) DOI: 10.1371/journal.pgen.1000534

PS: did you want these as separate items to be voted on? I had a second tool in here but took it out and can either edit this post with others or add new answers.

Possible language for paper (editable by others): Resources for exploring the biomedical literature may offer leads for further investigation. GRAIL (Gene Relationships Among Implicated Loci) combines the regional assessment of genes in a SNP location with text-mining of research papers about those genes to highlight possible significant terms.

ADD COMMENT
2
Entering edit mode

Mary - I think there are at least a dozen tool/database that need to be mentioned for the post-GWAS analysis. How about adding a table with tool/db with a one line description of details ?

ADD REPLY
0
Entering edit mode

thank you for your answer!! I think that it is better if you put it on another post. Moreover, I have read your blog post, but it would be nice if you can also put a few lines here to explain why this method could be used to characterize the function of a snp associated with a disease and discovered through a GWAS.

ADD REPLY
0
Entering edit mode

Yeah, I wasn't sure how much text to add at this point, or wait for additional guidance on the structure. But I'll edit this item with language specific for GRAIL (which will of course be open for editing by others as well as this develops).

ADD REPLY
0
Entering edit mode

Mary - I think there are at least a dozen tool/database that need to be mentioned for the post-GWAS analaysis - a one line description of each of the tools in a table will do the trick. How about adding a table with such details ?

ADD REPLY
0
Entering edit mode

@Khader: that sounds like an interesting way to go. They are also very popular with readers, I know. I was thinking in my head about what people new to this really need to know, and they even need to be told about dbSNP. Very obvious to us, but really not to everyone. Wasn't sure how far to take this...

ADD REPLY
0
Entering edit mode

I thought a bit about this, and maybe it is better if you also write this into the WikiGenes directly. That way you will have the opportunity to read the draft personally and to understand what is missing there and where this should be included. We can continue the discussion on both places, BioStar and WikiGenes; maybe Biostar is a better place to make the list of tools and WikiGenes is better to write the text collaboratively.

ADD REPLY
0
Entering edit mode

@Khader: the table is a good idea. Let's create a table in the wiki?

ADD REPLY
0
Entering edit mode

well, so far this is the proposal that I like the most. For me it is a good addition to the paper.

ADD REPLY
0
Entering edit mode

@Giovanni: So where do we start ? We have variants in 2 major categories- coding and non coding variant. We need different tools for these two analysis pipelines (please see the answer I posted regarding my ideas that I would like to incorporate)

ADD REPLY
0
Entering edit mode

Finally I read through current version. I am working on a work-flow diagram currently, I will add it here before publishing on the wiki. I will be editing the section "Database for SNP functional annotation and Computational approach to predict function".

I think the current version of table is a verbose list of tools: we need to classify them into different groups according to the application.

ADD REPLY
0
Entering edit mode

What is the significance of "A role for Digital Life Information" in this proposed post-GWAS paper ? IMHO, it is a totally out of context material for the manuscript.

ADD REPLY
2
Entering edit mode
14.0 years ago

I would like to add a paragraph regarding methods that can be used for the analysis of variants. I am focusing on two pipelines that enable "location based analysis of variants" from GWAS. I am thinking to add two pipelines that can be used for location based analysis of coding and non-coding SNPs.

From my post-GWAS analysis experience, a lot of information is missing in techniques that rely on database mapping. The users should be aware of the context of their gene (for example ORF, transcript, alternate splicing events) and perform the analysis.

ADD COMMENT
0
Entering edit mode

thank you Khader; if you want an advice, write this also in the WikiGene page, if you don't know where to put it you can add it as a note in the Discussion page. I think that what you say is interesting and could be added to the manuscript, but I am not sure what you mean with 'location based analysis of variants'; maybe you should put an example.

ADD REPLY
1
Entering edit mode
14.0 years ago
Mary 11k

I was waiting to see of others had ideas on this, but I'll keep going with my suggestions then....

I was going to mention the Genome Graphs tool at UCSC. Here's a post I did about it, and here's the Genome Graphs User Guide. I think this would be helpful for people because they could put their variants on and it would allow them to explore all the context available--including promoter regions and other possible regulatory aspects (histone modifications, etc).

Could be combined with the Ensembl strategy, or could be separate.

ADD COMMENT
0
Entering edit mode

thanks for putting it in the text. I would move your paragraph before mine, at the beginning

ADD REPLY
1
Entering edit mode
14.0 years ago
Mary 11k

As the emphasis on that paper is cancer, I'm going to offer the BioMart at ICGC as a tool.

This reminds me of that question we had before on tools for cancer--maybe more of them should be brought into this as well. I'm going to leave that to others who might want to contribute.

ADD COMMENT
1
Entering edit mode

Well, I think if you have a variant you'd want to check and see if it was also identified in tumor vs non-tumor, or in a CNV section, etc.

ADD REPLY
0
Entering edit mode

mmm ok, put it in the wiki, even if it may be too general. I agree with you, I was expecting more answers.

ADD REPLY
1
Entering edit mode
14.0 years ago

I should also put my own answers.

The first thing to do to characterize a snp identified in a GWAS is to look at the information available for the SNP in other databases. For example, see if it is known to be involved in other diseases or other mechanisms; dbSNP, OMIM and the EBI's variants databases can be useful. A blast search for the region near the SNP, or an analysis using a genome browser as Mary said, can provide good information.

So, in the table I would put the basic tools: blast, dbSNP, etc.. What do you think?

ADD COMMENT
0
Entering edit mode

Well, I think so. If people are going to come to this paper wondering what to do with their novel variants, they need that too. In fact, I've had people come up to me with exactly that question--and their grasp of databases + tools can be quite limited.

ADD REPLY
0
Entering edit mode

I think the table should be categorized as annotation database and tools to assess the functional effects.

ADD REPLY
0
Entering edit mode

ok, do it.. however since we also added a paragraph about snp prioritization/candidate genes, we should include it in the title.

ADD REPLY
1
Entering edit mode
14.0 years ago

SNPs falling within coding sequences

I have found this paper describing an analysis of non-synonymous SNPs. They analyzed thousands of SNPs falling into a coding sequence and predicted their effect on the protein.

Burke DF, Worth CL, Priego E-M, et al. Genome bioinformatic analysis of nonsynonymous SNPs. BMC bioinformatics. 2007;8:301. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17708757

The manuscript is looking good for now with our additions. I have to go for today, see you tomorrow!

ADD COMMENT

Login before adding your answer.

Traffic: 1832 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6