|InterJournal Complex Systems, 104
|Manuscript Number: |
Submission Date: 971010
|Genetic Network Inference|
Subject(s): CX.3, BG.2, BG.14
Category: Brief Article
Given the explosion of data in molecular biology, in particular, those generated by the human and animal genome projects, we are now faced with the challenge of a functional genomics. Traditionally, molecular biology has focused on the study of individual genes. We must now consider experimental and computational approaches for determining the self-organizing principles that govern parallel interactions among large numbers of genes. Such an approach will be necessary for the generation of a comprehensive theory of biological development. Development, as well as other processes of phenotypic change, such as those occurring during the progress of disease, may be conceptualized as the dynamic output of a genetic program. We are currently using the Boolean network analogy as a model for the genetic network underlying the developmental process. Boolean networks exhibit dynamic properties similar to those of biological systems; these include self-organization and cycling. We have applied the network analogy experimentally through the use of reverse-transcription polymerase chain reaction (RT-PCR) to measure the expression (mRNA) of large numbers of genes over the course of central nervous system development. We have measured mRNA expression at nine developmental time points for 65 genes in rat cervical spinal cord and hippocampus. Clustering of these temporal gene expression patterns using the euclidean distance measure revealed groups of genes which may share inputs. Constraints on reverse-engineering algorithms will be necessary for limiting the number of hypothetical wiring diagrams for a genetic network. These constraints include the results of perturbations on baseline temporal gene expression patterns, as well as assumptions concerning the types of combinatorial rules expected of a biological network. The collection of temporal gene expression patterns and development of computational methods, such as clustering and reverse engineering algorithms for analyzing these data, may provide new hypotheses concerning the study of tissue regeneration and cancer. Further, this approach may contribute to new insights necessary for understanding the flow of genetic information which characterizes the developmental process.
|Submit referee report/comment|