Joaquín Dopazo talks with
ScienceWatch.com and answers a few questions about
this month's Emerging Research Front Paper in the field of
Article: FatiGO: a web tool for finding significant
associations of Gene Ontology terms with groups of
Authors: Al-Shahrour, F;Diaz-Uriarte, R;Dopazo,
Journal: BIOINFORMATICS, 20 (4): 578-580 MAR 1 2004
Addresses: Ctr Nacl Invest Oncol, Bioinformat Unit, Melchor
Fernandez Almagro 3, Madrid 28029, Spain.
Ctr Nacl Invest Oncol, Bioinformat Unit, Madrid 28029, Spain.
Why do you think your paper is highly
The paper describes a web-based bioinformatic software that helps one to
understand the functional implications of the genes selected in a genomic
experiment. Since its publication, there has been a continuous growth in
the demand for these type of tools. In the particular case of
FatiGO, this functionality was offered in a very
straightforward manner within a rigorous statistical framework.
Actually, the universe of web-based bioinformatics tools exhibits a
peculiar Darwinian dynamic. To be user-friendly and quick in its
calculations, as well as staying up and running permanently, constitutes a
real competitive advantage for any tool. These facts, along with its timely
publication in a moment in which there was a growing interest for the
functional profiling of genomic data in the field are, most probably, the
primary reasons for its popularity.
Does it describe a new discovery, methodology, or
synthesis of knowledge?
"...the development of diagnostic or prognostic tools
based in DNA microarrays is among the aspects with more
social projection derived from our research. "
This paper describes a testing methodology, nowadays known under the
generic name of "enrichment analysis" that provides information on the
functional roles carried out by a group of pre-selected genes (or
proteins). Perhaps, the most innovative aspect included in the publication
was the introduction of the concept of multiple testing in the framework of
statistical contrast, which allowed us to eliminate a considerable number
of false positives in the analysis.
Would you summarize the significance of your paper
in layman’s terms?
The method presented in the paper allows the conversion of the lists of
uninformative codes of gene identifiers obtained in genomic experiments
into their corresponding shared functional roles. In this way, a researcher
can quickly have a pretty clear idea of the functionalities of the cell
affected by an experiment and easily relate them to the experimental
condition studied (i.e., disease, drug administration, etc.).
How did you become involved in this research and
were any particular problems encountered along the way?
My main interest has revolved around systems biology problems. The
definition of the sets of functionally-related genes altered by an
experiment constitutes the first step in the knowledge of the relationships
among the genes that ultimately outline the cell functionality.
There is a cultural problem, still extended among many researchers, that
prevents them from understanding that genes do not operate alone, but in a
complex network of interactions.
The reductionist view of genes in isolation, related to functions, has been
demonstrated to be an oversimplification of the reality which cannot any
longer be maintained. This view, unfortunately, constitutes a particular
burden for the advancement of several fields in which systems biology
should play an eminent role.
Where do you see your research leading in the
In line with the systems' perspective, my interests are pursuant to
understanding the relationships among genes that define the functional
modules (e.g., pathways) in the cell, as well as the high-level
relationships among such modules. A correct understanding of the assembly
and relationships among the pieces of the system will pave the way toward
achieving an operation over its functionality and reverting perturbations,
such as diseased states, etc.
Do you foresee any social or political
implications for your research?
Although my work is quite far away from a translational use in biomedicine,
some aspects are of immediate application. For example, the development of
diagnostic or prognostic tools based on DNA microarrays derived from our
research is among the aspects offering more social projection.
Joaquín Dopazo, Ph.D.
Department of Bioinformatics and Genomics
Centro de Investigacion Principe Felipe (CIPF)
KEYWORDS: FatiGO; FALSE DISCOVERY RATE; EXPRESSION DATA-ANALYSIS.