About DIOPT and DIOPT-DIST

The identification of orthologs is commonly used for bioinforamtics activities such as data mining and establishing models for human diseases. Moreover, our group notes that researchers analyzing the results of screens performed at the Drosophila RNAi Screening Center (DRSC) frequently wish to identify mammalian orthologs of the fly genes that were "hits" (positive results) in their screens.

In helping DRSC screeners to identify orthologs using existing tools and algorithms, we recognized a need for a user-friendly approach to viewing and comparing ortholog predictions obtained using different tools and algorithms. This was our motivation in developing DIOPT. To facilitate identification of orthologs specifically of human disease-associated genes, we further developed DIOPT-DIST. Information about our approaches to development of both tools is summarized below.

Cite DIOPT or DIOPT-DIST

If you use DIOPT or DIOPT-DIST in your research, we ask that you please cite our paper:

Hu Y, Flockhart I, Vinayagam A, Bergwitz C, Berger B, Perrimon N, Mohr SE. An Integrative Approach to Ortholog Prediction for Disease-Focused and Other Functional Studies. BMC Bioinformatics. 2011 Aug 31;12(1):357.

The DIOPT Approach

Many tools have emerged to meet the need to identify orthologs. However, low coverage and heterogeneity of these tools present an obstacle to scientists who want to identify a one or a few highest-confidence orthologs for a given gene of interest or conversely, want to cast a wide net and follow up on all possible orthologs of a gene.

Our goal is to provide an easy-to-use resource that facilitates summary, comparison and access to various sources of ortholog predictions. DIOPT integrates human, mouse, fly, worm, zebrafish and yeast ortholog predictions made by Ensembl Compara, HomoloGene, Inparanoid, Isobase, OMA, orthoMCL, Phylome, RoundUp, and TreeFam. DIOPT lets users find ortholog pairs for a specified gene or genes identified by one, many or all of these published approaches. This provides a streamlined method for integration, comparison and access to orthology predictions originating from algorithms based on sequence homology, phylogenetic trees, and functional similarity. DIOPT calculates a simple score indicating the number of tools that support a given orthologous gene-pair relationship, as well as a weighted score based on functional assessment using high quality GO molecular function annotation of all fly-human orthologous pairs predicted by each tool. Differences in the algorithms used by each tool to predict orthologous relationship is one source of difference in the set of predictions made by one tool versus another. However, we also note that some of these differences might be attributable to use of different genome annotation releases used by some tools versus others, and that not all tools cover all of the species that we include in the DIOPT tool (see Tables 1,2 and 3).

DIOPT also displays protein and domain alignments, including percent amino acid identity, for predicted ortholog pairs. These should help you to identify the most appropriate matches among multiple possible orthologs.

The following summary figures and tables help to explain our approach and summarize the tools and algorithms included in DIOPT.

Figure 1: Summary of the DIOPT approach to integration of results from multiple ortholog prediction tools and algorithms. In green, tools based on sequence alignment. In purple, tools based on evolutionary relationships. In orange, a tool that incorporates protein-protein interaction network data into ortholog predictions.

Table 1: Summary Information and Publications for the Tools Integrated in DIOPT

* DIOPT weights are based on the mean semantic similarity of high quality GO molecular function annotation of all fly-human orthologous pairs predicted by each tool.
   BBH, Best Blast Hits
   RSD, Reciprocal Smallest Distance
   PPI, Protein-Protein Interactions

Table 2A: Genome Release Information for the Tools Integrated in DIOPT

Table 2B: Additional Information About Genome Releases

Table 3. Maximum DIOPT score for each orthologous relationship

The DIOPT-DIST Approach

Facilitating the identification of orthologs between a model organism and humans is of particular relevance to genes associated with human diseases. Keeping in mind that researchers visiting or otherwise using resources from the DRSC are often interested to identify orthologs of disease-associated genes, including genes more recently identified through genome-wide association studies (GWAS), we decided to take DIOPT results a step further by linking them up with disease associations in curated resources.

DIOPT-DIST is an online-searchable tool that maps gene-disease relationships from the NCBI Online Menedlian Inheritance in Man (OMIM) database and GWAS datasets contained in the GWAS catalog to genes in the C. elegans, Drosophila, mouse, S. cerevisiae, and zebrafish genomes using DIOPT orthologous predictions. Disease terms were extracted from OMIM and GWAS ftp files and were categorized with MeSH headings. 2 additional categories were added for the terms out of the scope of MeSH disease annotation such as disease risk factors and traits (table 4).

The user may search 1.) the related human diseases by a list of genes from model organisms; 2.) the corresponding genes in model organisms by disease term, disease category or IDs from OMIM. Both gene/locus OMIM IDs (starts with * or + on the OMIM website eg. 100650) and disease phenotype OMIM IDs (starts with # or % on the OMIM website eg. 610251) can be searched at DIOPT-DIST (table 5).

Figure 2: The DIOPT-DIST Approach

Table 4. Disease Categories

Table 5. DIOPT-DIST result page, one example of OMIM terms

• 100650 is an example of an OMIM ID for gene/locus terms, starting with a "+" at the OMIM website (http://omim.org/entry/100650?search=100650&highlight=100650 ).
• 610251 is an example of an OMIM ID for disease phenotype terms, starting with a "#" at the OMIM website (http://omim.org/entry/610251?search=610251&highlight=610251 ).

• 610251 is the OMIM ID for the phenotype.
• Brackets, "{ }", indicate mutations that contribute to susceptibility to multi-factorial disorders or to susceptibility to infection.
• "(3)" means the molecular basis of the disorder is known.

For more information about OMIM annotation, please go to the OMIM help page (http://omim.org/help/faq).

If you have questions, suggestions or comments on DIOPT please contact our informatics staff.