Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours?

Amin Zollanvari, Nancy L. Saccone, Laura J. Bierut, Marco F. Ramoni, Gil Alterovitz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Gene expression and genome wide association data have provided researchers the opportunity to study many complex traits and diseases. When designing prognostic and predictive models capable of phenotypic classification in this area, significant reduction of dimensionality through stringent filtering and/or feature selection is often deemed imperative. Here, this work challenges this presumption through both theoretical and empirical analysis. This work demonstrates that by a proper compromise between structure of the selected model and the number of features, one is able to achieve better performance even in large dimensionality. The inclusion of many genes/variants in the classification rules can help shed new light on the analysis of complex traitstraits that are typically determined by many causal variants with small effect size.

Original languageEnglish
Title of host publicationProceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
Pages3573-3576
Number of pages4
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 - Boston, MA, United States
Duration: Aug 30 2011Sep 3 2011

Other

Other33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
CountryUnited States
CityBoston, MA
Period8/30/119/3/11

Fingerprint

Genes
Gene expression
Feature extraction
Research Personnel
Association reactions
Genome
Gene Expression

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Biomedical Engineering
  • Health Informatics

Cite this

Zollanvari, A., Saccone, N. L., Bierut, L. J., Ramoni, M. F., & Alterovitz, G. (2011). Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours? In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (pp. 3573-3576). [6090596] https://doi.org/10.1109/IEMBS.2011.6090596

Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours? / Zollanvari, Amin; Saccone, Nancy L.; Bierut, Laura J.; Ramoni, Marco F.; Alterovitz, Gil.

Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2011. p. 3573-3576 6090596.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zollanvari, A, Saccone, NL, Bierut, LJ, Ramoni, MF & Alterovitz, G 2011, Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours? in Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS., 6090596, pp. 3573-3576, 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011, Boston, MA, United States, 8/30/11. https://doi.org/10.1109/IEMBS.2011.6090596
Zollanvari A, Saccone NL, Bierut LJ, Ramoni MF, Alterovitz G. Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours? In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2011. p. 3573-3576. 6090596 https://doi.org/10.1109/IEMBS.2011.6090596
Zollanvari, Amin ; Saccone, Nancy L. ; Bierut, Laura J. ; Ramoni, Marco F. ; Alterovitz, Gil. / Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours?. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2011. pp. 3573-3576
@inproceedings{618d9e1e9aff4f53bf93fca3a50f6075,
title = "Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours?",
abstract = "Gene expression and genome wide association data have provided researchers the opportunity to study many complex traits and diseases. When designing prognostic and predictive models capable of phenotypic classification in this area, significant reduction of dimensionality through stringent filtering and/or feature selection is often deemed imperative. Here, this work challenges this presumption through both theoretical and empirical analysis. This work demonstrates that by a proper compromise between structure of the selected model and the number of features, one is able to achieve better performance even in large dimensionality. The inclusion of many genes/variants in the classification rules can help shed new light on the analysis of complex traitstraits that are typically determined by many causal variants with small effect size.",
author = "Amin Zollanvari and Saccone, {Nancy L.} and Bierut, {Laura J.} and Ramoni, {Marco F.} and Gil Alterovitz",
year = "2011",
doi = "10.1109/IEMBS.2011.6090596",
language = "English",
isbn = "9781424441211",
pages = "3573--3576",
booktitle = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",

}

TY - GEN

T1 - Is the reduction of dimensionality to a small number of features always necessary in constructing predictive models for analysis of complex diseases or behaviours?

AU - Zollanvari, Amin

AU - Saccone, Nancy L.

AU - Bierut, Laura J.

AU - Ramoni, Marco F.

AU - Alterovitz, Gil

PY - 2011

Y1 - 2011

N2 - Gene expression and genome wide association data have provided researchers the opportunity to study many complex traits and diseases. When designing prognostic and predictive models capable of phenotypic classification in this area, significant reduction of dimensionality through stringent filtering and/or feature selection is often deemed imperative. Here, this work challenges this presumption through both theoretical and empirical analysis. This work demonstrates that by a proper compromise between structure of the selected model and the number of features, one is able to achieve better performance even in large dimensionality. The inclusion of many genes/variants in the classification rules can help shed new light on the analysis of complex traitstraits that are typically determined by many causal variants with small effect size.

AB - Gene expression and genome wide association data have provided researchers the opportunity to study many complex traits and diseases. When designing prognostic and predictive models capable of phenotypic classification in this area, significant reduction of dimensionality through stringent filtering and/or feature selection is often deemed imperative. Here, this work challenges this presumption through both theoretical and empirical analysis. This work demonstrates that by a proper compromise between structure of the selected model and the number of features, one is able to achieve better performance even in large dimensionality. The inclusion of many genes/variants in the classification rules can help shed new light on the analysis of complex traitstraits that are typically determined by many causal variants with small effect size.

UR - http://www.scopus.com/inward/record.url?scp=84055199163&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84055199163&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2011.6090596

DO - 10.1109/IEMBS.2011.6090596

M3 - Conference contribution

C2 - 22255111

AN - SCOPUS:84055199163

SN - 9781424441211

SP - 3573

EP - 3576

BT - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

ER -