The PhenogramViz team illustrates how they analyze and visualize gene-phenotype relationships

Video Tip of the Week: PhenogramViz for evaluating phenotypes and CNVs

As I’ve mentioned before, once I start looking over some new tools I’m often led to others in the same arena that offer related but different features. That’s what happened when I looked at the Proband iPad app for human pedigrees. I noted that they are using important community standards, and I decided to follow those threads a bit. That led me to last week’s tip, the Human Phenotype Ontology (HPO).

HPO has been around for a while and I’ve been aware of it, but this recent re-investigation made me realize how mature it has become, and I was impressed with the amount of adoption there’s been in the genomics community in the big projects. But it also led me to some new tools that I hadn’t encountered before. This week’s tip highlights PhenogramViz–combining my appreciation for controlled vocabularies, standards, and data visualization.

The PhenogramViz team illustrates how they analyze and visualize gene-phenotype relationships

The PhenogramViz team illustrates how they analyze and visualize gene-phenotype relationships

Here’s now the PhenogramViz team describes their tool:

A tool that automatically analyses and visualizes gene-to-phenotype relations for a set of genes affected by CNV of a patient and a set of HPO-terms representing the symptoms of said patient. The tool makes full use of the cross-species phenotype ontology “uberpheno” (see here).

So if you have a patient with copy-number variation issues in their genome, you may be able to use this tool to lead to the genes in that CNV segment that convey certain phenotypes. So the goal–as stated in their paper linked below–is to assist with the clinical interpretation of the genome alterations.

The additional layer of this effort that I find useful is that they use another ontology to take this even further for supporting information. They employ the “Uberpheno” cross-species phenotype ontology to find further details in model organisms.

I’ll let you get a sense of how this works with one of their tutorial videos from their YouTube channel. They have others too–which will help you with different aspects on everything from installation to analyses. I’ll embed the one that shows how you start with a list of patient symptoms or phenotypes, then loading the CNVs or genes, then from the results list you can simply click for graphical representations of the gene-phenotype relationships. Then with the Cytoscape tools you can interact with the “phenograms” in more detail. There’s no sound, you can read the guidance in the callouts.

The videos include some abbreviations–like HPO. That’s why I talked last week about the Human Phenotype Ontology. I was prepping you for this one.  And in another video (Prioritization of pathogenic CNVs) they reference the scoring strategies, which you will find need further explanation in their paper linked below (Journal of Medical Genetics one). I would spend some time looking over how the scoring and ranking happens to understand what’s shown.

Although the focus of this is using the data for human diagnosis, I think it could also help researchers to choose more appropriate animal model for further testing. There are lots of complaints about the unsuitability of animal models for a range of subjects–but refining those choices would also be a huge benefit. Saving resources by helping to choose the right animal model would be another worthwhile use of this tool.

Check out PhenogramViz as a bridge between genomic segments and possible phenotypes. You can try it yourself with sample files they have available on their landing page.

Quick links:

PhenogramViz: http://compbio.charite.de/contao/index.php/phenoviz.html

Cytoscape: http://cytoscape.org/

References:

Köhler, S., Doelken, S., Mungall, C., Bauer, S., Firth, H., Bailleul-Forestier, I., Black, G., Brown, D., Brudno, M., Campbell, J., FitzPatrick, D., Eppig, J., Jackson, A., Freson, K., Girdea, M., Helbig, I., Hurst, J., Jahn, J., Jackson, L., Kelly, A., Ledbetter, D., Mansour, S., Martin, C., Moss, C., Mumford, A., Ouwehand, W., Park, S., Riggs, E., Scott, R., Sisodiya, S., Vooren, S., Wapner, R., Wilkie, A., Wright, C., Vulto-van Silfhout, A., Leeuw, N., de Vries, B., Washingthon, N., Smith, C., Westerfield, M., Schofield, P., Ruef, B., Gkoutos, G., Haendel, M., Smedley, D., Lewis, S., & Robinson, P. (2013). The Human Phenotype Ontology project: linking molecular biology and disease through phenotype data Nucleic Acids Research, 42 (D1) DOI: 10.1093/nar/gkt1026

Köhler S., Doelken S.C., Ruef B.J., Bauer S., Washington N., Westerfield M., Gkoutos G., Schofield P., Smedley D. & Lewis S.E. & (2013). Construction and accessibility of a cross-species phenotype ontology along with gene annotations for biomedical research., F1000Research, PMID: http://www.ncbi.nlm.nih.gov/pubmed/24358873

Köhler, S., Schoeneberg, U., Czeschik, J., Doelken, S., Hehir-Kwa, J., Ibn-Salem, J., Mungall, C., Smedley, D., Haendel, M., & Robinson, P. (2014). Clinical interpretation of CNVs with cross-species phenotype data Journal of Medical Genetics, 51 (11), 766-772 DOI: 10.1136/jmedgenet-2014-102633

Shannon P., Markiel A., Ozier O., Baliga N.S., Wang J.T., Ramage D., Amin N., Schwikowski B. & Ideker T. Cytoscape: a software environment for integrated models of biomolecular interaction networks., Genome research, PMID: http://www.ncbi.nlm.nih.gov/pubmed/14597658