Limits...
Artificial neural networks for linkage analysis of quantitative gene expression phenotypes and evaluation of gene x gene interactions.

Liu Y, Duan W, Paschall J, Saccone NL - BMC Proc (2007)

Bottom Line: However some ANN results were noisy, and our attempts to use cross-validated training to avoid overtraining and thereby improve results were only partially successful.Potential interactions between loci with high-ranked weight measures were also evaluated, with the resulting patterns suggesting existence of both synergistic and antagonistic effects between loci.However, for the approach implemented here, optimizing the ANNs and obtaining stable results remains challenging.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Genetics, Washington University School of Medicine, 4566 Scott Avenue, Box 8232, St, Louis, Missouri 63110, USA. yliua@artsci.wustl.edu

ABSTRACT

Background: Using single-nucleotide polymorphism (SNP) genotypes and selected gene expression phenotypes from 14 CEPH (Centre d'Etude du Polymorphisme Humain) pedigrees provided for Genetic Analysis Workshop 15 (GAW15), we analyzed quantitative traits with artificial neural networks (ANNs). Our goals were to identify individual linkage signals and examine gene x gene interactions. First, we used classical multipoint methods to identify phenotypes having nominal linkage evidence at two or more loci. ANNs were then applied to sib-pair identity-by-descent (IBD) allele sharing across the genome as input variables and squared trait sums and differences for the sib pairs as output variables. The weights of the trained networks were analyzed to assess the linkage evidence at each locus as well as potential interactions between them.

Results: Loci identified by classical linkage analysis could also be identified by our ANN analysis. However some ANN results were noisy, and our attempts to use cross-validated training to avoid overtraining and thereby improve results were only partially successful. Potential interactions between loci with high-ranked weight measures were also evaluated, with the resulting patterns suggesting existence of both synergistic and antagonistic effects between loci.

Conclusion: Our results suggest that ANNs can serve as a useful method to analyze quantitative traits and are a potential tool for detecting gene x gene interactions. However, for the approach implemented here, optimizing the ANNs and obtaining stable results remains challenging.

No MeSH data available.


Related in: MedlinePlus

ANN analysis of 210910_s_at (ZP3), using Map 2 and all sib pairs.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2367483&req=5

Figure 2: ANN analysis of 210910_s_at (ZP3), using Map 2 and all sib pairs.

Mentions: ANN results for 210910_s_at (ZP3) using Map 2 are given in Figure 2. The highest peak identified by the ANN is consistent with the strongest linkage peak from both H-E and VC analyses, and the secondary peak on chromosome 14 matches one of the additional notable linkage peaks. Correlation results are given in Table 2. Interestingly, although chromosome 15 (the second best signal from traditional analysis) was not detected by the ANN method (Fig. 2), the interaction analysis highlighted a strong correlation with the chromosome 7 locus.


Artificial neural networks for linkage analysis of quantitative gene expression phenotypes and evaluation of gene x gene interactions.

Liu Y, Duan W, Paschall J, Saccone NL - BMC Proc (2007)

ANN analysis of 210910_s_at (ZP3), using Map 2 and all sib pairs.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2367483&req=5

Figure 2: ANN analysis of 210910_s_at (ZP3), using Map 2 and all sib pairs.
Mentions: ANN results for 210910_s_at (ZP3) using Map 2 are given in Figure 2. The highest peak identified by the ANN is consistent with the strongest linkage peak from both H-E and VC analyses, and the secondary peak on chromosome 14 matches one of the additional notable linkage peaks. Correlation results are given in Table 2. Interestingly, although chromosome 15 (the second best signal from traditional analysis) was not detected by the ANN method (Fig. 2), the interaction analysis highlighted a strong correlation with the chromosome 7 locus.

Bottom Line: However some ANN results were noisy, and our attempts to use cross-validated training to avoid overtraining and thereby improve results were only partially successful.Potential interactions between loci with high-ranked weight measures were also evaluated, with the resulting patterns suggesting existence of both synergistic and antagonistic effects between loci.However, for the approach implemented here, optimizing the ANNs and obtaining stable results remains challenging.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Genetics, Washington University School of Medicine, 4566 Scott Avenue, Box 8232, St, Louis, Missouri 63110, USA. yliua@artsci.wustl.edu

ABSTRACT

Background: Using single-nucleotide polymorphism (SNP) genotypes and selected gene expression phenotypes from 14 CEPH (Centre d'Etude du Polymorphisme Humain) pedigrees provided for Genetic Analysis Workshop 15 (GAW15), we analyzed quantitative traits with artificial neural networks (ANNs). Our goals were to identify individual linkage signals and examine gene x gene interactions. First, we used classical multipoint methods to identify phenotypes having nominal linkage evidence at two or more loci. ANNs were then applied to sib-pair identity-by-descent (IBD) allele sharing across the genome as input variables and squared trait sums and differences for the sib pairs as output variables. The weights of the trained networks were analyzed to assess the linkage evidence at each locus as well as potential interactions between them.

Results: Loci identified by classical linkage analysis could also be identified by our ANN analysis. However some ANN results were noisy, and our attempts to use cross-validated training to avoid overtraining and thereby improve results were only partially successful. Potential interactions between loci with high-ranked weight measures were also evaluated, with the resulting patterns suggesting existence of both synergistic and antagonistic effects between loci.

Conclusion: Our results suggest that ANNs can serve as a useful method to analyze quantitative traits and are a potential tool for detecting gene x gene interactions. However, for the approach implemented here, optimizing the ANNs and obtaining stable results remains challenging.

No MeSH data available.


Related in: MedlinePlus