Dear All,
I am trying to see which CpG sites (with its associated genes) are involved in particular pathways and diseases, and get an overview of the functions of these genes.
Currently, I have tried to import my dataset (>800k CpG sites total) which shows the following: 1) each CpG site as the ID, 2) p-value, 3) q-value, 4) fold change and 5) difference. My data sets are quite large with >200,000 CpG sites (the row limit of IPA) - is there a way to import a file this large?
I have also tried importing a file with more specific CpG sites of around 1000 CpG sites but it is not being mapped properly by IPA as I have 0 mapped sites due to errors or possibly I am using the wrong template (i.e. not expression data)?
I think the errors are coming from my formatting in my excel file to IPA, where either the headings are incorrect and the way I am assigning each header/observation is incorrect i.e. I think I set my Identifier as Illumina (which is what I used to get my CpG methylation data), but I do not know what other options I can choose instead of this. IPA also showed errors first with 'no IDs matched to particular genes',and then with 'removing fold change between 1 and -1'.
In summary, I would really appreciate any tips/guidance with uploading CpG methylation data into IPA.
Thank you very much.
What kind of species are we talking about?
You could use GREAT to associate your CpG sites to genes.
Also, if you have a generally random distribution over the genome, you can decide to only import the genes which you think are biologically relevant differences only, since the background would than be basically all annotated genes. Furthermore, just import the gene list. The direction of DNA methylation in not so informative for IPA, since you never know what the relation is of a specific differential methylated site and the associated gene, if there is any.
Use ensembl IDs, or entrez ID. Use biomart to convert your gene names.
Hope this helps.
Jorke
First you select CpG sites which are differentially methylated based on p-value or fold-change. Now map CpG sites to the nearest genes based on intergenic distance cut-off. Get these gene names and provide it to IPA. You may try tool suggested by Mr. Kamstra
Thank you very much Mr. Kamstra and Dr. Muley!
I am looking at human samples from infant cord blood lymphocytes.
So far, I have a smaller list of CpG sites that are relevant based on p-values and fold-change (as what Dr.Muley suggested).
Mr. Kamstra, I have associated CpG sites with genes using GenomeStudio, but would I have to use biomart to convert these gene names to ensemble IDs or entrez IDs?
Thank you for your help!
You can use BiomaRt or uniprot id mapper to convert ensembl ids to entrez
Dear, I want to use IPA also to show the relations between differentially methylated genes. Can I import the gene list only, without another information? How IPA can analyse the pathway if it is activated or inhibated without any value (only with the list of gene)? Thank you
- Badges
- Science method