Biomedical subjects

Lesley Collins

Publications and source records attributed to Lesley Collins.

4 recordsLinked to original sources

PubMed2006

Bayesian learning of sparse gene regulatory networks.

Differential equations (DEs) have been the most widespread formalism for gene regulatory network (GRN) modeling, as they offer natural interpretation of biological processes, easy elucidation of gene relationships, and the capability of using efficient parameter estimation methods. However, an important limitation of DEs is their requirement of O(d(2)) parameters where d is the number of genes modeled, which often causes over-parameterization for large d, leading to the over-fitting of data and dense parameter sets that are hard to interpret. This paper presents the first effort to address the over-parameterization problem by applying the sparse Bayesian learning (SBL) method to sparsify the GRN model of DEs. SBL operates on the parsimony principle, with the objective to reduce the number of effective parameters by driving the redundant parameters to zero. The resulting sparse parameter set offers three important advantages for GRN inference: first, the inferred GRNs are more plausible, since the biological counterparts are known to be sparse; second, gene relationships can be more easily elucidated from sparse sets than from dense sets; and third, the solutions become more optimal and consistent, due to the reduction in the volume of solution space. Experiments are conducted on the yeast Saccharomyces cerevisiae time-series gene expression data, in which known regulatory events related to the cell cycle G1/S phase are reliably reproduced.

Bayes Theorem↗

PubMed2005

Proceedings of the SMBE Tri-National Young Investigators' Workshop 2005. Investigating the intron recognition mechanism in eukaryotes.

Recent studies indicate that many introns, as well as the complex spliceosomal mechanism to remove them, were present early in eukaryotic evolution. This study examines intron and exon characteristics from annotations of whole genomes to investigate the intron recognition mechanism. Exon definition uses the exon as the unit of recognition, placing length constraints on the exon but not on the intron (allowing it a greater range of lengths). In contrast, intron definition uses the intron itself as the unit of recognition and thus removes constraints on internal exon length forced by the use of an exon definition mechanism. Thus, intron and exon lengths within a genome can reflect the constraints imposed by its splicing. This study shows that it is possible firstly to recover valid intron and exon information from genome annotation. We then compare internal intron and exon information from a range of eukaryotic genomes and investigate possible evolutionary length constraints on introns and exons and how they can impact on the intron recognition mechanism. Results indicate that exon definition-based mechanisms may predominate in vertebrates although the exact system in fish is expected to show some differences with the better characterized system from mammals. We also raise the possibility that the last common ancestor of plants and animals contained some type of exon definition and that this mechanism was replaced in some genes and lineages by intron definition, possibly as a result of intron loss and/or intron shortening.

Alternative Splicing↗

PubMed2005

Complex spliceosomal organization ancestral to extant eukaryotes.

In higher eukaryotes, introns are spliced out of protein-coding mRNAs by the spliceosome, a massive complex comprising five non-coding RNAs (ncRNAs) and about 200 proteins. By comparing the differences between spliceosomal proteins from many basal eukaryotic lineages, it is possible to infer properties of the splicing system in the last common ancestor of extant eukaryotes, the eukaryotic ancestor. We begin with the hypothesis that, similar to intron length (that appears to have increased in multicellular eukaryotes), the spliceosome has increased in complexity throughout eukaryotic evolution. However, examination of the distribution of spliceosomal components indicates that not only was a spliceosome present in the eukaryotic ancestor but it also contained most of the key components found in today's eukaryotes. All the small nuclear ribonucleoproteins (snRNPs) protein components are likely to have been present, as well as many splicing-related proteins. Both major and trans-splicing are likely to have been present, and the spliceosome had already formed links with other cellular processes such as transcription and capping. However, there is no evidence as yet to suggest that minor (U12-dependent) splicing was present in the eukaryotic ancestor. Although the last common ancestor of extant eukaryotes appears to show much of the molecular complexity seen today, we do not, from this work, infer anything of the properties of the earlier "first eukaryote."

Eukaryotic Cells↗

PubMed2005

A hybrid genetic algorithm and expectation maximization method for global gene trajectory clustering.

Clustering time-course gene expression data (gene trajectories) is an important step towards solving the complex problem of gene regulatory network modeling and discovery as it significantly reduces the dimensionality of the gene space required for analysis. Traditional clustering methods that perform hill-climbing from randomly initialized cluster centers are prone to produce inconsistent and sub-optimal cluster solutions over different runs. This paper introduces a novel method that hybridizes genetic algorithm (GA) and expectation maximization algorithms (EM) for clustering gene trajectories with the mixtures of multiple linear regression models (MLRs), with the objective of improving the global optimality and consistency of the clustering performance. The proposed method is applied to cluster the human fibroblasts and the yeast time-course gene expression data based on their trajectory similarities. It outperforms the standard EM method significantly in terms of both clustering accuracy and consistency. The biological implications of the improved clustering performance are demonstrated.

Algorithms↗