A Rule-Based Framework for Gene Regulation Pathways Discovery Page: 3 of 4
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
A rule-based framework for gene regulation pathways discovery
Bartosz Wilczynski
Lawrence Livermore National Laboratory
Livermore, CA, USA
Warsaw University,
Warsaw, Poland
Andriy Kryshtafovych
Lawrence Livermore National Laboratory
Livermore, CA, USA
Jan Komorowski
Linnaeus Centre for Bioinformatics
Uppsala, Sweden
Abstract
We present novel approach to discover the rules that gov-
ern gene regulation mechanisms. The method is based on
supervised machine learning and is designed to reveal rela-
tionships between transcription factors and gene promoters.
As the representation of the gene regulatory circuit we have
chosen a special form of IF-THEN rules associating cer-
tain features (a generalized idea of a Transcription Factor
Binding Site) in gene promoters with specific gene expres-
sion profiles.1 Introduction
Understanding of the gene regulation mechanisms is cur-
rently one of the most important tasks for molecular bi-
ology. New genome sequences become available every
month, but we are still far from being able to reliably
and consistently unravel the corresponding gene regulatory
pathways. As the number of known genomes is growing
and the average size of a genome sequence dataset is large,
we expect the method of pathway discovery to be general,
automated, and efficient. We describe our approach, based
on the assumption that genome sequences, with known gene
positions, together with the gene expression data, are suf-
ficient to find all the interactions between genes and their
transcription factors.Torgeir Hvidsten
Lawrence Livermore National Laboratory
Livermore, CA, USA
Linnaeus Centre for Bioinformatics
Uppsala, Sweden
Lisa Stubbs
Lawrence Livermore National Laboratory
Livermore, CA, USA
Krzysztof Fidelis
Lawrence Livermore National Laboratory
Livermore, CA, USA
2 Algorithm outline
A given set of genes, with expression levels measured
under specific conditions, is used as input to our procedure.
We start with gene expression profile clustering and then
iteratively perform the following steps:
* Searching for significant features in the gene regula-
tory regions
* Inducing rules from genes clustered by expression and
a feature set
* Evaluating the rules
* Refining the feature set until no further progress can be
achieved.
3 Local similarity of gene expression profiles
We will assume that the genes that are co-regulated by a
common transcription factor show significant correlation in
their expression profiles. Many authors (like [6, 5]) search
for the transcription factor binding sites in the regulatory
regions of genes with similar expression profiles. Some
transcription factors, however, may bind only under certain
conditions. Indeed, for example genes from baker's yeast
demonstrate expression correlation only during a part of the
yeast cell cycle. This leads us to a modified approach to
gene expression clustering. We consider groups of genes
that show very high correlation only in a part of the expres-
sion profile.
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Wilczynski, B; Hvidsten, T; Kryshtafovych, A; Stubbs, L; Komorowski, J & Fidelis, K. A Rule-Based Framework for Gene Regulation Pathways Discovery, article, July 21, 2003; Livermore, California. (https://digital.library.unt.edu/ark:/67531/metadc1407864/m1/3/: accessed June 3, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.