Limits...
Autoregressive higher-order hidden Markov models: exploiting local chromosomal dependencies in the analysis of tumor expression profiles.

Seifert M, Abou-El-Ardat K, Friedrich B, Klink B, Deutsch A - PLoS ONE (2014)

Bottom Line: The performance benefit is attributed to the simultaneous usage of higher-order state-transitions in combination with autoregressive emissions.This benefit could not be reached by using each of these two features independently.An implementation is available under www.jstacs.de/index.php/ARHMM.

View Article: PubMed Central - PubMed

Affiliation: Center for Information Services and High Performance Computing, Dresden University of Technology, Dresden, Germany.

ABSTRACT
Changes in gene expression programs play a central role in cancer. Chromosomal aberrations such as deletions, duplications and translocations of DNA segments can lead to highly significant positive correlations of gene expression levels of neighboring genes. This should be utilized to improve the analysis of tumor expression profiles. Here, we develop a novel model class of autoregressive higher-order Hidden Markov Models (HMMs) that carefully exploit local data-dependent chromosomal dependencies to improve the identification of differentially expressed genes in tumor. Autoregressive higher-order HMMs overcome generally existing limitations of standard first-order HMMs in the modeling of dependencies between genes in close chromosomal proximity by the simultaneous usage of higher-order state-transitions and autoregressive emissions as novel model features. We apply autoregressive higher-order HMMs to the analysis of breast cancer and glioma gene expression data and perform in-depth model evaluation studies. We find that autoregressive higher-order HMMs clearly improve the identification of overexpressed genes with underlying gene copy number duplications in breast cancer in comparison to mixture models, standard first- and higher-order HMMs, and other related methods. The performance benefit is attributed to the simultaneous usage of higher-order state-transitions in combination with autoregressive emissions. This benefit could not be reached by using each of these two features independently. We also find that autoregressive higher-order HMMs are better able to identify differentially expressed genes in tumors independent of the underlying gene copy number status in comparison to the majority of related methods. This is further supported by the identification of well-known and of previously unreported hotspots of differential expression in glioblastomas demonstrating the efficacy of autoregressive higher-order HMMs for the analysis of individual tumor expression profiles. Moreover, we reveal interesting novel details of systematic alterations of gene expression levels in known cancer signaling pathways distinguishing oligodendrogliomas, astrocytomas and glioblastomas. An implementation is available under www.jstacs.de/index.php/ARHMM.

Show MeSH

Related in: MedlinePlus

Systematic characterization of the most discriminative signaling pathways distinguishing different types of gliomas.a) Characteristic view on the most discriminative pathways between oligodendrogliomas, astrocytomas and glioblastomas at the level of the top 300 overexpressed genes in Figure 5. b) Selected gene-based view on the most discriminative signaling pathways shown in a). The Venn diagrams show pathway-specific overlaps of overexpressed genes between the different types of gliomas. The strong overlap of genes between the different types of gliomas indicates the presence of common core sets of affected genes. These pathway-specific core gene sets are further extended towards the glioma with the greatest number of overexpressed genes. The corresponding genes are summarized in Table 2.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4067306&req=5

pone-0100295-g006: Systematic characterization of the most discriminative signaling pathways distinguishing different types of gliomas.a) Characteristic view on the most discriminative pathways between oligodendrogliomas, astrocytomas and glioblastomas at the level of the top 300 overexpressed genes in Figure 5. b) Selected gene-based view on the most discriminative signaling pathways shown in a). The Venn diagrams show pathway-specific overlaps of overexpressed genes between the different types of gliomas. The strong overlap of genes between the different types of gliomas indicates the presence of common core sets of affected genes. These pathway-specific core gene sets are further extended towards the glioma with the greatest number of overexpressed genes. The corresponding genes are summarized in Table 2.

Mentions: Finally, we analyze the most discriminative pathways between oligodendrogliomas, astrocytomas and glioblastomas shown in Figure 5 at the level of single genes to investigate whether the different types of gliomas utilize the same sets of genes to alter pathway activities. Corresponding pathway-specific barplots and Venn diagrams for the top 300 overexpressed genes are shown in Figure 6. The Venn diagrams clearly indicate that the different types of gliomas mainly utilize pathway-specific common core sets of affected genes. These pathway-specific core gene sets are further extended towards the type of glioma with the greatest number of overexpressed genes. Since this observation might be of potential relevance for the development of tumor-specific markers and future treatment strategies, we have summarized the underlying genes and their corresponding pathway memberships in the most discriminative cancer signaling pathways in Table 2. Interestingly, it is important to note that of the 41 listed genes 17 genes are playing a role in at least two pathways and that 12 of these genes are involved in three pathways. Among the genes involved in three pathways, the combination of ECM-Receptor interaction, PI3K-Akt signaling and Focal adhesion is strongly overrepresented, which underlines the complexity and the interplay of pathway alterations in tumors.


Autoregressive higher-order hidden Markov models: exploiting local chromosomal dependencies in the analysis of tumor expression profiles.

Seifert M, Abou-El-Ardat K, Friedrich B, Klink B, Deutsch A - PLoS ONE (2014)

Systematic characterization of the most discriminative signaling pathways distinguishing different types of gliomas.a) Characteristic view on the most discriminative pathways between oligodendrogliomas, astrocytomas and glioblastomas at the level of the top 300 overexpressed genes in Figure 5. b) Selected gene-based view on the most discriminative signaling pathways shown in a). The Venn diagrams show pathway-specific overlaps of overexpressed genes between the different types of gliomas. The strong overlap of genes between the different types of gliomas indicates the presence of common core sets of affected genes. These pathway-specific core gene sets are further extended towards the glioma with the greatest number of overexpressed genes. The corresponding genes are summarized in Table 2.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4067306&req=5

pone-0100295-g006: Systematic characterization of the most discriminative signaling pathways distinguishing different types of gliomas.a) Characteristic view on the most discriminative pathways between oligodendrogliomas, astrocytomas and glioblastomas at the level of the top 300 overexpressed genes in Figure 5. b) Selected gene-based view on the most discriminative signaling pathways shown in a). The Venn diagrams show pathway-specific overlaps of overexpressed genes between the different types of gliomas. The strong overlap of genes between the different types of gliomas indicates the presence of common core sets of affected genes. These pathway-specific core gene sets are further extended towards the glioma with the greatest number of overexpressed genes. The corresponding genes are summarized in Table 2.
Mentions: Finally, we analyze the most discriminative pathways between oligodendrogliomas, astrocytomas and glioblastomas shown in Figure 5 at the level of single genes to investigate whether the different types of gliomas utilize the same sets of genes to alter pathway activities. Corresponding pathway-specific barplots and Venn diagrams for the top 300 overexpressed genes are shown in Figure 6. The Venn diagrams clearly indicate that the different types of gliomas mainly utilize pathway-specific common core sets of affected genes. These pathway-specific core gene sets are further extended towards the type of glioma with the greatest number of overexpressed genes. Since this observation might be of potential relevance for the development of tumor-specific markers and future treatment strategies, we have summarized the underlying genes and their corresponding pathway memberships in the most discriminative cancer signaling pathways in Table 2. Interestingly, it is important to note that of the 41 listed genes 17 genes are playing a role in at least two pathways and that 12 of these genes are involved in three pathways. Among the genes involved in three pathways, the combination of ECM-Receptor interaction, PI3K-Akt signaling and Focal adhesion is strongly overrepresented, which underlines the complexity and the interplay of pathway alterations in tumors.

Bottom Line: The performance benefit is attributed to the simultaneous usage of higher-order state-transitions in combination with autoregressive emissions.This benefit could not be reached by using each of these two features independently.An implementation is available under www.jstacs.de/index.php/ARHMM.

View Article: PubMed Central - PubMed

Affiliation: Center for Information Services and High Performance Computing, Dresden University of Technology, Dresden, Germany.

ABSTRACT
Changes in gene expression programs play a central role in cancer. Chromosomal aberrations such as deletions, duplications and translocations of DNA segments can lead to highly significant positive correlations of gene expression levels of neighboring genes. This should be utilized to improve the analysis of tumor expression profiles. Here, we develop a novel model class of autoregressive higher-order Hidden Markov Models (HMMs) that carefully exploit local data-dependent chromosomal dependencies to improve the identification of differentially expressed genes in tumor. Autoregressive higher-order HMMs overcome generally existing limitations of standard first-order HMMs in the modeling of dependencies between genes in close chromosomal proximity by the simultaneous usage of higher-order state-transitions and autoregressive emissions as novel model features. We apply autoregressive higher-order HMMs to the analysis of breast cancer and glioma gene expression data and perform in-depth model evaluation studies. We find that autoregressive higher-order HMMs clearly improve the identification of overexpressed genes with underlying gene copy number duplications in breast cancer in comparison to mixture models, standard first- and higher-order HMMs, and other related methods. The performance benefit is attributed to the simultaneous usage of higher-order state-transitions in combination with autoregressive emissions. This benefit could not be reached by using each of these two features independently. We also find that autoregressive higher-order HMMs are better able to identify differentially expressed genes in tumors independent of the underlying gene copy number status in comparison to the majority of related methods. This is further supported by the identification of well-known and of previously unreported hotspots of differential expression in glioblastomas demonstrating the efficacy of autoregressive higher-order HMMs for the analysis of individual tumor expression profiles. Moreover, we reveal interesting novel details of systematic alterations of gene expression levels in known cancer signaling pathways distinguishing oligodendrogliomas, astrocytomas and glioblastomas. An implementation is available under www.jstacs.de/index.php/ARHMM.

Show MeSH
Related in: MedlinePlus