How To Analyse Epigenetics Data Using Bioinformatics Tools?

Question

[Mandar]

How to analyse epigenetics data using bioinformatics tools? can anybody tell me where can I get the epigenomic data like DNA methylome? any database?

EDIT: I have global methylation pattern in Drosophila which we obtained using 5 methyl cytosine antibody coupled cDNA microarray based approach. I have a gene names and their methylation value.can anybody tell me how to analyse such data?

[brentp]

If you edit your question to be more specific, you'll get more helpful replies. Do you want to look at Human? At Cancer? Are you interested in a set of genes?

[Michael]

Hi, you seem to like single-sentence questions (this Q is in fact 2 in one and is no exception). You also have been asked to provide further details for many of your questions, but have hardly ever done so. I would like to ask you to try to be more specific next time, to help us to give you an answer, be in line with the standards of this site, and to show respect to experts investing a multiple of the time you have invested to type this sloppy 30sec. question to answer it. This question is at the verge of being closed!

As general guidance for a question of a beginner, I recommend that it consists of at least one or two paragraphs detailing your work, what you have tried before, the data and biological question you are facing.

[onguma382]

it's funny I just sent email minutes ago saying the same thing like yours. Yes indeed the specific question is showing how the person making an afford to get the answer :)

[Sean Davis]

Given the comments and your replies, I think your best bet here is to find a local collaborator to work with you on getting started. There is a real temptation to approach bioinformatics as a "how do I do it" problem, but often the more important questions such as "what do I want to know" are harder to define. A person with some experience in the field can get you a long way in answering both.

[Michael]

Be careful, I am very suspicious that this whole question is made up. First it is about "two cancers" now it is drosophila? Fortunately, it is not a requirement that a question cannot be entirely made up, but it explains, why the level of detail cannot be higher.

[Eric Normandeau]

In the presence of so little clarity and the absence of any real improvements, I would consider closing this question and potentially add a long comment about how to write better questions in the future.

Answer 1

[Michael]

There are two questions here: How to analyse epigenetics data using bioinformatics tools? This is too broad to be answered precisely. You need to specify what kind of data you have at hand and what the biological question is.

Edit: If looking for a review paper, a google search for "methylation data" will do the trick.

Edit2: If you really have one methylation value per gene, then the analysis most likely ends here. If your question is real, I have the impression that this is not a state-of-the-art approach, one value per gene is imo merely useless, when you can have 10-1bp resolution with modern techniques, cDNA arrays were new about 15 years ago. So in conclusion it might be best to not further analyse these data.

Database of epigenomics data: The ENCODE project has possibly created the largest collection of such data. You might find this resource useful: http://www.genome.gov/27551933 There are also ENCODE tracks in the UCSC genome browser.

Answer 2

[Devon Ryan]

You can download a lot of data from GEO or SRA. How to analyse things will depend entirely on the nature of the data and the question you have. "how it looks means nature of the data" doesn't parse.

[Mandar]

I want to analyse DNA methylome in two different cancers,how to go about it?

[Sean Davis]

Please be more specific. What cancers? What datasets (give accessions or platform type)?

[Michael]

Sometimes, this kind of response makes me think there are people making up problems, that do not exist....

Answer 3

[ugly.betty77]

The word 'epigenetics' is scientifically meaningless and is a fad as explained by Mark Ptashne many times.

Faddish Stuff: Epigenetics and the Inheritance of Acquired Characteristics On the use of the word ‘epigenetic’

Scientific basis for epigenome projects has been questioned by Ptashne, Hobert and Davidson. Questions over the scientific basis of epigenome project

FWIW.

[Istvan Albert]

Biostar is not an online forum or mailing list but a Q&A. Back and forth discussions are not well supported nor are encouraged/desired.

This post above breaks the rule of not being an answer yet it is posted in the category of an answer to the question. It should be probably added as a comment but even as such it may not be all that appropriate since it does not help answering or clarifying the main question nor the answer.

[ugly.betty77]

Let us say a flat-earther asks here what steps to take to make sure he does not fall off after reaching the end of the earth. How can telling him that earth is not flat not be a proper answer? In fact not telling him gives legitimacy to his question and misleads others, who also read the Q/A.

Epigenetics is junk science. The term itself is not precisely defined, and the claims related to it are not scientifically valid. If you have other evidence, please feel free to show it to me.

[Honey]

I cannot resist myself, This is a true observation about Epigentics field. Even if you see literature there is no consistency how you will call differential methylation, Whatever suits you take that parameter/s. It is so superficial science. There are few people they can publish in high standard journals ( even crap) rest is a history.

[Michael]

This seems to be highly controversial and the statements a minority opinion. Also, I do not see how this is helpful as an answer to state that the topic is 'non-sense'.

[ugly.betty77]

Mark Ptashne, Hobert and Eric Davidson are 'minority opinion'? Geez !

Answer 4

[Charles Warden]

I would agree with the concerns about the comment being pretty general, but I do happen to have a list of tools for DNA methylation analysis (including my own algorithm, COHCAP, which has been published in Nucleic Acids Research relatively recently). Hopefully, this can help:

http://cdwscience.blogspot.com/2013/03/bioinformatics-101-dna-methylation.html

[Mandar]

I have global methylation pattern in Drosophila which we obtained using 5 methyl cytosine antibody coupled cDNA microarray based approach. I have a gene names and their methylation value.can anybody tell me how to analyse such data?

[Charles Warden]

MeDIP is just a special case of ChIP-chip (where you could utilize an algorithm like MAT).

However, MeDIP / ChIP-chip is really meant for tiling arrays. cDNA arrays are typically used for gene expression analysis. So, interpretation of such results may be difficult.