Question: How to graphically display multidimensional categorical data
0
gravatar for nadarajanvelu
3.8 years ago by
nadarajanvelu10 wrote:

Hi,

I have a purely categorical dataset and want display in the form of an appropriate graph. The data is in the form of CSV. The data has information about enzyme names, biological pathway, organisms and the chromosome numbers they belong to. I want to display the chromosome to which these enzymes belong to by pathway and organism. Please let me know the best way to display this data. I know R and D3.js. Not sure whether these tools are good for this type of data or not.

Thanks

graphical display R • 1.0k views
ADD COMMENTlink modified 5 days ago by Biostar ♦♦ 20 • written 3.8 years ago by nadarajanvelu10

What would you want to understand from the graph after it is drawn?

ADD REPLYlink written 3.8 years ago by Chris Evelo9.9k

To see whether any particular pathway enzymes prefer a chromosome, and if so whether it is conserved across species.

ADD REPLYlink written 3.8 years ago by nadarajanvelu10
2

In that case why not create a dataset that contains the chromosome number for each gene and do a pathway overrepresentation analysis for each chromosome? 

ADD REPLYlink written 3.8 years ago by Chris Evelo9.9k

Please let me know if there are any open source software to do such analysis.

Thanks

ADD REPLYlink written 3.8 years ago by nadarajanvelu10

Yes, there are severall. There are R packages that can do the numerical part and Go-Elite can do that too. The advantage of these could be that they can use the gene-pathway assignments that you already have. Personally I would prefer to do it in PathVisio, provided sufficient pathways for your species of interest are available. Note you need standard over representation analysis (z-score based on a matching a criterium) and not what is usually called gene set enrichment analysis since that will consider you chromosome values as a level scale.

ADD REPLYlink written 3.8 years ago by Chris Evelo9.9k
0
gravatar for Lynxoid
3.8 years ago by
Lynxoid210
Pleasanton, CA
Lynxoid210 wrote:

Parallel Sets is good for finding trends in categorical data.

ADD COMMENTlink written 3.8 years ago by Lynxoid210
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 832 users visited in the last hour