|
Evogene's core competence relies on its unique proprietary computational gene discovery platform – the "ATHLETE" (Agro Trait Harvest LEads TEchnology). Athlete uses vast amounts of available genomic data (mostly public) to rapidly reach a reliable limited list of candidate key genes with high relevance to a target trait of choice. Allegorically, the Athlete platform could be viewed as a "machine" that is able to choose 50-100 lottery tickets from amongst hundreds of thousands of tickets, with the high likelihood that the winning ticket will be included among them.
Key characteristics of the "ATHLETE" platform
Reliable - Results from model and target plant validation in our various projects have shown the reliability and efficiency of the platform in discovering key genes.
Novel - The platform is comprised of unique algorithmic tools and data mining concepts.
Flexible - The platform can be utilized for different traits and crops.
Rapid - A discovery round takes 6 to 9 months.
The ATHLETE platform utilizes Compugen’s (NASDAQ: CGEN) recognized state-of-the-art computational "LEADS" platform, which was designed for human genome analysis by Compugen and subsequently adapted and enriched by Evogene for plant applications. The LEADS platform has been licensed to leading pharmaceutical companies such as Pfizer, Novartis, and Abbott Laboratories and is scientifically recognized by leading scientific journals such as Science and Nature Biotechnology.
ATHLETE is actually a toolbox of computational tools that are combined and adapted according to the project's needs. These tools make it possible to refine, structure, profile, correlate, compare, mine and prioritize vast amounts of genomic data in order to reach a limited list of key genes. A gene discovery process utilizing the ATHLETE platform lasts 6 to 9 months, initiating from hundreds of thousands of genomic sequences to produce a limited list of tens of key genes relevant to the trait. The process includes the following phases:
Trait Understanding
Understanding the nature of the trait and developing several approaches and strategies to tackle the problem is key to generating a heterogeneous population of genes. A team of professional experts is formed, including plant biologists, biochemists, breeders, growers and others. This team, led by our project leader, builds different strategies to tackle the trait, which are translated later into the computational database and data mining.
Database Assembly
This phase enables the building of a wide, rich, reliable, annotated and easy-to-analyze database that is comprised of available genomic data from various crops, such as expressed sequences (mRNAs and ESTs), genomic sequences, expression data from microarray experiments and other relevant information. Our database assembly is comprised of a toolbox of gene refining, structuring, annotation and analysis tools that enable us to construct a tailored database for each project. Currently our database is comprised of over 70 plant species and more than 400K proprietary gene clusters formed by over 8M expressed sequences.
Gene refining and structuring tools make it possible to reliably detect splice variants, generating understanding of various phenotypic outcomes of a single gene. Several additional proprietary analysis tools are key to enriching our data base and to conducting efficient comparative genomics.
Data Mining
Several mining approaches and proprietary algorithmic tools are used to filter out and prioritize candidate genes during this final stage. The final stage includes review of outcome and prioritization by our team of scientific advisors, as previously described. Data mining utilizes a toolbox of algorithmic and statistical tools that are used according to each project's specific needs. Multidimensional keyword based queries combine the strategies decided upon in the "Trait Understanding" phase and the systematic "Evologs" and "Gene Babylon" data assembly previously conducted.
Other tools include our "Hook" system for identifying novel genes based on features of known relevant genes associated to the trait and systems biology algorithms, such as our proprietary "Path D Way" tool for linking unknown genes to known pathways.
The outcome of this phase is a list of 50- 100 prioritized and annotated candidate genes with high relevance to the trait to further undergo validation and prioritization.
|
 |
More Technologies:
Overview
Gene Validation and Prioritization
|
