A closed-form solution for transcription factor activity estimation using network component analysis

Amina Noor, Aitzaz Ahmad, Bilal Wajid, Erchin Serpedin, Mohamed Nounou, Hazem Nounou

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

2 Citations (Scopus)

Abstract

Non-iterative network component analysis (NINCA), proposed by Jacklin at.al, employs convex optimization methods to estimate the transcription factor control strengths and transcription factor activities. While NINCA provides good estimation accuracy and higher consistency, the costly optimization routine used therein renders a high computational complexity. This correspondence presents a closed form solution to estimate the connectivity matrix which is tens of times faster, and provides similar accuracy and consistency, thus making the closed form NINCA (CFNINCA) algorithm useful for large data sets encountered in practice. The proposed solution is assessed for accuracy and consistency using synthetic and yeast cell cycle data sets by comparing with the existing state-of-the-art algorithms. The robustness of the algorithm to the possible inaccuracies in prior information is also analyzed and it is observed that CFNINCA and NINCA are much more robust to erroneous prior information as compared to FastNCA.

Original languageEnglish
Title of host publicationAlgorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings
PublisherSpringer Verlag
Pages196-207
Number of pages12
Volume8542 LNBI
ISBN (Print)9783319079523
DOIs
Publication statusPublished - 2014
Event1st International Conference on Algorithms for Computational Biology, AlCoB 2014 - Tarragona, Spain
Duration: 1 Jul 20143 Jul 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8542 LNBI
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other1st International Conference on Algorithms for Computational Biology, AlCoB 2014
CountrySpain
CityTarragona
Period1/7/143/7/14

Fingerprint

Network components
Transcription factors
Transcription Factor
Closed-form Solution
Prior Information
Closed-form
Algorithm Analysis
Convex optimization
Cell Cycle
Convex Optimization
Large Data Sets
Yeast
Estimate
Optimization Methods
Computational complexity
Computational Complexity
Connectivity
Correspondence
Cells
Robustness

Keywords

  • convex optimization
  • Gene Regulatory Network
  • transcription factor activity

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Noor, A., Ahmad, A., Wajid, B., Serpedin, E., Nounou, M., & Nounou, H. (2014). A closed-form solution for transcription factor activity estimation using network component analysis. In Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings (Vol. 8542 LNBI, pp. 196-207). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8542 LNBI). Springer Verlag. https://doi.org/10.1007/978-3-319-07953-0_16

A closed-form solution for transcription factor activity estimation using network component analysis. / Noor, Amina; Ahmad, Aitzaz; Wajid, Bilal; Serpedin, Erchin; Nounou, Mohamed; Nounou, Hazem.

Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings. Vol. 8542 LNBI Springer Verlag, 2014. p. 196-207 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8542 LNBI).

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

Noor, A, Ahmad, A, Wajid, B, Serpedin, E, Nounou, M & Nounou, H 2014, A closed-form solution for transcription factor activity estimation using network component analysis. in Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings. vol. 8542 LNBI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8542 LNBI, Springer Verlag, pp. 196-207, 1st International Conference on Algorithms for Computational Biology, AlCoB 2014, Tarragona, Spain, 1/7/14. https://doi.org/10.1007/978-3-319-07953-0_16
Noor A, Ahmad A, Wajid B, Serpedin E, Nounou M, Nounou H. A closed-form solution for transcription factor activity estimation using network component analysis. In Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings. Vol. 8542 LNBI. Springer Verlag. 2014. p. 196-207. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-07953-0_16
Noor, Amina ; Ahmad, Aitzaz ; Wajid, Bilal ; Serpedin, Erchin ; Nounou, Mohamed ; Nounou, Hazem. / A closed-form solution for transcription factor activity estimation using network component analysis. Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings. Vol. 8542 LNBI Springer Verlag, 2014. pp. 196-207 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{866bbf0db9674284894a498009ca0400,
title = "A closed-form solution for transcription factor activity estimation using network component analysis",
abstract = "Non-iterative network component analysis (NINCA), proposed by Jacklin at.al, employs convex optimization methods to estimate the transcription factor control strengths and transcription factor activities. While NINCA provides good estimation accuracy and higher consistency, the costly optimization routine used therein renders a high computational complexity. This correspondence presents a closed form solution to estimate the connectivity matrix which is tens of times faster, and provides similar accuracy and consistency, thus making the closed form NINCA (CFNINCA) algorithm useful for large data sets encountered in practice. The proposed solution is assessed for accuracy and consistency using synthetic and yeast cell cycle data sets by comparing with the existing state-of-the-art algorithms. The robustness of the algorithm to the possible inaccuracies in prior information is also analyzed and it is observed that CFNINCA and NINCA are much more robust to erroneous prior information as compared to FastNCA.",
keywords = "convex optimization, Gene Regulatory Network, transcription factor activity",
author = "Amina Noor and Aitzaz Ahmad and Bilal Wajid and Erchin Serpedin and Mohamed Nounou and Hazem Nounou",
year = "2014",
doi = "10.1007/978-3-319-07953-0_16",
language = "English",
isbn = "9783319079523",
volume = "8542 LNBI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "196--207",
booktitle = "Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings",

}

TY - GEN

T1 - A closed-form solution for transcription factor activity estimation using network component analysis

AU - Noor, Amina

AU - Ahmad, Aitzaz

AU - Wajid, Bilal

AU - Serpedin, Erchin

AU - Nounou, Mohamed

AU - Nounou, Hazem

PY - 2014

Y1 - 2014

N2 - Non-iterative network component analysis (NINCA), proposed by Jacklin at.al, employs convex optimization methods to estimate the transcription factor control strengths and transcription factor activities. While NINCA provides good estimation accuracy and higher consistency, the costly optimization routine used therein renders a high computational complexity. This correspondence presents a closed form solution to estimate the connectivity matrix which is tens of times faster, and provides similar accuracy and consistency, thus making the closed form NINCA (CFNINCA) algorithm useful for large data sets encountered in practice. The proposed solution is assessed for accuracy and consistency using synthetic and yeast cell cycle data sets by comparing with the existing state-of-the-art algorithms. The robustness of the algorithm to the possible inaccuracies in prior information is also analyzed and it is observed that CFNINCA and NINCA are much more robust to erroneous prior information as compared to FastNCA.

AB - Non-iterative network component analysis (NINCA), proposed by Jacklin at.al, employs convex optimization methods to estimate the transcription factor control strengths and transcription factor activities. While NINCA provides good estimation accuracy and higher consistency, the costly optimization routine used therein renders a high computational complexity. This correspondence presents a closed form solution to estimate the connectivity matrix which is tens of times faster, and provides similar accuracy and consistency, thus making the closed form NINCA (CFNINCA) algorithm useful for large data sets encountered in practice. The proposed solution is assessed for accuracy and consistency using synthetic and yeast cell cycle data sets by comparing with the existing state-of-the-art algorithms. The robustness of the algorithm to the possible inaccuracies in prior information is also analyzed and it is observed that CFNINCA and NINCA are much more robust to erroneous prior information as compared to FastNCA.

KW - convex optimization

KW - Gene Regulatory Network

KW - transcription factor activity

UR - http://www.scopus.com/inward/record.url?scp=84904012937&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904012937&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-07953-0_16

DO - 10.1007/978-3-319-07953-0_16

M3 - Conference contribution

SN - 9783319079523

VL - 8542 LNBI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 196

EP - 207

BT - Algorithms for Computational Biology - First International Conference, AlCoB 2014, Proceedings

PB - Springer Verlag

ER -