DES-TOMATO: A Knowledge Exploration System Focused On Tomato Species

Adil Salhi,Salim Bougouffa,Robert Hoehndorf,Rozaimi Razali,Maxat Kulmanov,Aleksandar Radovanovic,Vladimir B Bajic,Benoit Marchand,Magbubah Essack,Mitchell J L Morton,Sónia Negrão,Mark Tester

doi:10.1038/s41598-017-05448-0

Adil Salhi, Salim Bougouffa + Show 10 more

Open Access

https://doi.org/10.1038/s41598-017-05448-0

Copy DOI

Abstract

Tomato is the most economically important horticultural crop used as a model to study plant biology and particularly fruit development. Knowledge obtained from tomato research initiated improvements in tomato and, being transferrable to other such economically important crops, has led to a surge of tomato-related research and published literature. We developed DES-TOMATO knowledgebase (KB) for exploration of information related to tomato. Information exploration is enabled through terms from 26 dictionaries and combination of these terms. To illustrate the utility of DES-TOMATO, we provide several examples how one can efficiently use this KB to retrieve known or potentially novel information. DES-TOMATO is free for academic and nonprofit users and can be accessed at http://cbrc.kaust.edu.sa/des_tomato/, using any of the mainstream web browsers, including Firefox, Safari and Chrome.

Highlights

In an attempt to provide an independent assessment of the quality of associations identified by KB, we evaluated the quality of the gene pairs extracted by the KB by comparing them to their functional similarity, where functions of the genes are obtained from an independent data source
We hypothesize that a strong correlation between our extracted associations between gene pairs and their functional similarity is reflective of the quality of the data in DES-TOMATO and its analysis approach
We propose that a correct association between two genes in DES-TOMATO will generally be reflected by the two genes’ sharing similar GO annotations, some gene pairs may be associated in a manner not reflected by GO term similarities

Summary

Introduction

We demonstrate the effectiveness of DES-TOMATO in finding useful associations by presenting four case studies These examples demonstrate how users can, with ease and speed, identify putative candidate genes, build a network of gene regulation for a specific trait, generate topic-specific hypotheses and explore enriched pathways. To our knowledge, this is the only KB derived through literature text-mining that has a comprehensive information exploration capabilities dedicated to the Lycopersicon section of the Solanaceae. DES was originally developed by VBB and AR and subsequently improved in various ways

Results

Discussion

Conclusion