Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer

Wei Hu

doi:doi:10.11648/j.bio.20160402.12

| Peer-Reviewed

Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer

Wei Hu

Published in American Journal of Bioscience and Bioengineering (Volume 4, Issue 2)

Received: 3 May 2016 Accepted: Published: 4 May 2016

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Various gene signatures of chemosensitivity in breast cancer have been identified. When used to build predictors of have chemosensitivity, many of them have their prediction accuracy around 80%. Identifying gene signatures to build high accuracy such predictors is a prerequisite for their clinical tests and applications. To elucidate the importance of each individual gene in a signature is another pressing need before such signature could be tested in clinical settings. In this study, Genetic Algorithms (GAs) and Sparse Logistic Regression (SLR) were employed to identify two signatures. The first had 28 probe sets selected by GA from the top 65 probe sets that were highly overexpressed between pathologic compete response (pCR) and residual disease (RD) and was used to build a SLR predictor of pCR (SLR-28). The second had 86 probe sets (Notch-86) selected by GA from Notch signaling pathway and was used to develop a SLR predictor of pCR (SLR-Notch-86). These two predictors tested on a training set (n=81) and validation set (n=52) had very precise predictions measured by accuracy, specificity, sensitivity, positive predictive value and negative predictive value with their corresponding P value all zero. Furthermore, these two predictors discovered 12 important genes in the 28 probe set signature and 14 important genes in the Notch-86 signature. Our two signatures produced superior performance over a signature in a previous study, demonstrating the potential of GA and SLR in identifying robust gene signatures in chemo response prediction in breast cancer.

Published in	American Journal of Bioscience and Bioengineering (Volume 4, Issue 2)
DOI	10.11648/j.bio.20160402.12
Page(s)	26-33
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Genetic Algorithm, Gene Signature, Breast Cancer, Sparse Logistic Regression, Predictor, Chemosensitivity

References

[1]	Hess, K. R., Anderson, K., Symmans, W. F., Valero, V., Ibrahim, N., Mejia, J. A., Booser, D., Theriault, R. L., Buzdar, A. U., Dempsey, P. J., Rouzier, R., Sneige, N., Ross, J. S., Vidaurre, T., Gómez, H. L., Hortobagyi, G. N. and Pusztai, L. (2006) Pharmacogenomic predictor of sensitivity to preoperative chemotherapy with paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide in breast cancer, J. ClinOncol, Vol. 24, pp. 4236–4244.
[2]	Chanrion M, Negre V, Fontaine H, Salvetat N, Bibeau F, Mac Grogan G, Mauriac L, Katsaros D, Molina F, Theillet C, Darbon JM. (2008) A gene expression signature that can predict the recurrence of tamoxifen-treated primary breast cancer. Clin Cancer Res. 14(6): 1744-52.
[3]	Linke SP, Bremer TM, Herold CD, Sauter G, Diamond C. (2006) A multimarker model to predict outcome in tamoxifen-treated breast cancer patients. Clin Cancer Res. 12(4): 1175-83.
[4]	P. E. Lønning, S. Knappskog, V. Staalesen, R. Chrisanthar & J. R. Lillehaug (2007) Breast cancer prognostication and prediction in the postgenomic era, Annals of Oncology 18: 1293–1306.
[5]	Folgueira MA, Carraro DM, Brentani H, Patrão DF, Barbosa EM, Netto MM, Caldeira JR, Katayama ML, Soares FA, Oliveira CT, Reis LF, Kaiano JH, Camargo LP, Vêncio RZ, Snitcovsky IM, Makdissi FB, e Silva PJ, Góes JC, Brentani MM. (2005) Gene expression profile associated with response to doxorubicin-based therapy in breast cancer. Clin Cancer Res. 11(20): 7434-43.
[6]	Holly K. Dressman, Christopher Hans6, Andrea Bild, John A. Olson, Eric Rosen, P. Kelly Marcom, Vlayka B. Liotcheva, Ellen L. Jones, Zeljko Vujaskovic, Jeffrey Marks, Mark W. Dewhirst, Mike West, Joseph R. Nevins and Kimberly Blackwell (2006) Gene Expression Profiles of Multiple Breast Cancer Phenotypes and Response to Neoadjuvant Chemotherapy, Clinical Cancer Research Vol. 12, 819-826.
[7]	Olaf Thuerigen, Andreas Schneeweiss, Grischa Toedt, Patrick Warnat, Meinhard Hahn, Heidi Kramer, Benedikt Brors, Christian Rudlowski, Axel Benner, Florian Schuetz, Bjoern Tews, Roland Eils, Hans-Peter Sinn, Christof Sohn, Peter Lichter (2006) Gene Expression Signature Predicting Pathologic Complete Response With Gemcitabine, Epirubicin, and Docetaxel in Primary Breast Cancer, Journal of Clinical Oncology, Vol 24, No 12 pp. 1839-1845.
[8]	Goldstein NS, Decker D, Severson D et al. (2007) Molecular classification system identifies invasive breast carcinoma patients who are most likely and those who are least likely to achieve a complete pathologic response after neoadjuvant chemotherapy, Cancer 110: 1687–1696.
[9]	Chang, J. C. et al. (2003) Gene expression profiling for the prediction of therapeutic response to docetaxel in patients with breast cancer. Lancet 362, 362–369.
[10]	Gianni, L. et al. (2005) Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer, J. Clin. Oncol. 23, 7265–7277.
[11]	Hannemann, J. et al. (2005) Changes in gene expression associated with response to neoadjuvant chemotherapy in breast cancer. J. Clin. Oncol. 23, 3331–3342.
[12]	Singer CF, Klinglmüller F, Stratmann R, Staudigl C, Fink-Retter A, Gschwantler D, et al. (2013) Response Prediction to Neoadjuvant Chemotherapy: Comparison between Pre-Therapeutic Gene Expression Profiles and In Vitro Chemosensitivity Assay. PLoS ONE 8(6): e66573.
[13]	Torsten NielsenEmail author, Brett Wallden, Carl Schaper, Sean Ferree, Shuzhen Liu, Dongxia Gao, Garrett Barry, Naeem Dowidar, Malini Maysuria and James Storhoff (2014) Analytical validation of the PAM50-based Prosigna Breast Cancer Prognostic Gene Signature Assay and nCounter Analysis System using formalin-fixed paraffin-embedded breast tumor specimens, BMC Cancer 201414: 177.
[14]	Brian David Lehmann, Yan Ding, Daniel Joseph Viox, Ming Jiang, Yi Zheng, Wang Liao, Xi Chen, Wei and Yajun Yi (2015) Evaluation of public cancer datasets and signatures identifies TP53 mutant signatures with robust prognostic and predictive value, BMC Cancer 201515: 179.
[15]	Prat A, Lluch A, Albanell J, Barry WT, Fan C, Chacón JI, Parker JS, Calvo L, Plazaola A, Arcusa A, Seguí-Palmer MA, Burgues O, Ribelles N, Rodriguez-Lescure A, Guerrero A, Ruiz-Borrego M, Munarriz B, López JA, Adamo B, Cheang MC, Li Y, Hu Z, Gulley ML, Vidal MJ, Pitcher BN, Liu MC, Citron ML, Ellis MJ, Mardis E, Vickery T, Hudis CA, Winer EP, Carey LA, Caballero R, Carrasco E, Martín M, Perou CM, Alba E (2014) Predicting response and survival in chemotherapy-treated triple-negative breast cancer, Br J Cancer. 2014 Oct 14; 111(8): 1532-41.
[16]	Tibshirani, R. (1996) Regression shrinkage and selection via the lasso, J. Royal Statist. Soc. B, Vol. 58, pp. 267–288.
[17]	Shevade, S. K. and Keerthi, S. S. (2003) A simple and efficient algorithm for gene selection using sparse logistic regression, Bioinformatics, Vol. 19, pp. 2246–2253.
[18]	Cawley, G. C. and Talbot, L. C. (2006) Gene selection in cancer classification using sparse logistic regression with bayesian regularization, Bioinformatics, Vol. 22, pp. 2348–2355.
[19]	Brennan, K. and Anthony Brown, M. C. (2003) Is there a role for Notch signaling in human breast cancer? Breast Cancer Res, 5(2), 69-75.
[20]	Stylianou, S., Clarke, R. B. et al. (2006) Activation of notch signaling in human breast cancer, Cancer Research, 66, 1517-1525.
[21]	Xiaolin Huang Li Wang He Zhang Haibo Wang Xiaoping Zhao Guanxiang Qian Jifan Hu Shengfang Ge Xianqun Fan (2012) Therapeutic Efficacy by Targeting Correction of Notch1-Induced Aberrants in Uveal Tumors, PLoS ONE, 7(8): e44301.
[22]	GEArray, O. Human notch signaling pathway microarray. http://www.sabiosciences.com/gene_array_product/HTML/OHS-059.html
[23]	Chia SK, Wykoff CC, Watson PH, Han C, Leek RD, Pastorek J, Gatter KC, Ratcliffe P, Harris AL. (2001) Prognostic significance of a novel hypoxia-regulated marker, carbonic anhydrase IX, in invasive breast carcinoma. J Clin Oncol. 19(16): 3660-8.
[24]	Hussain SA, Ganesan R, Reynolds G, Gross L, Stevens A, Pastorek J, Murray PG, Perunovic B, Anwar MS, Billingham L, James ND, Spooner D, Poole CJ, Rea DW, Palmer DH (2007) Hypoxia-regulated carbonic anhydrase IX expression is associated with poor survival in patients with invasive breast cancer, Br J Cancer. 96(1): 104-9.
[25]	Daniel H. Barnett, Shubin Sheng, Tze Howe Charn, Abdul Waheed, William S. Sly, Chin-Yo Lin, Edison T. Liu and Benita S. Katzenellenbogen (2008) Estrogen Receptor Regulation of Carbonic Anhydrase XII through a Distal Enhancer in Breast Cancer, Cancer Research, 68, 3505.
[26]	Martin C Abba, Yuhui Hu, Carla C Levy, Sally Gaddis, Frances S Kittrell, Yun Zhang, Jamal Hill, Reid P, Daniel Medina, Powel H Brown and C Marcelo Aldaz (2008) Transcriptomic signature of Bexarotene (Rexinoid LGD1069) on mammary gland from three transgenic mouse mammary cancer models, BMC Medical Genomics, 1: 40.
[27]	Antonio Isidoro 1, Enrique Casado 2, Andrés Redondo 2, Paloma Acebo 1, Enrique Espinosa 2, Andrés M. Alonso 4, Paloma Cejas 2, David Hardisson 3, Juan A. Fresno Vara 2, Cristobal Belda-Iniesta 2, Manuel González-Barón 2 and José M. Cuezva (2005) Breast carcinomas fulfill the Warburg hypothesis and provide metabolic markers of cancer prognosis, Carcinogenesis, 26 (12): 2095-2104.
[28]	Rouzier R, Rajan R, et al. (2005) Microtubule associated protein tau is a predictive marker and modulator of response to paclitaxel-containing preoperative chemotherapy in breast cancer. Proc Natl Acad Sci U S A, 102, 8315-8320.
[29]	Ou YH, Pei-Han Chung PH, et al. (2007) The candidate tumor suppressor BTG3 is a transcriptional target of p53 that inhibits E2F1. The EMBO Journal, 26, 3968–3980.
[30]	Majid S, Dar AA, Ahmad AE, Hirata H, Kawakami K, Shahryari V, Saini S, Tanaka Y, Dahiya AV, Khatri G, Dahiya R. (2009) BTG3 tumor suppressor gene promoter demethylation, histone modification and cell cycle arrest by genistein in renal cancer, Carcinogenesis, 30 (4): 662-70.
[31]	Cheng YC1, Lin TY, Shieh SY. (2013) Candidate tumor suppressor BTG3 maintains genomic stability by promoting Lys63-linked ubiquitination and activation of the checkpoint kinase CHK1, Proc Natl Acad Sci U S A. 110 (15): 5993-8.
[32]	Goldstein NS, Decker D, Severson D et al. (2007) Molecular classification system identifies invasive breast carcinoma patients who are most likely and those who are least likely to achieve a complete pathologic response after neoadjuvant, chemotherapy. Cancer, 110: 1687–1696.

Cite This Article

Plain Text BibTeX RIS

APA Style

Wei Hu. (2016). Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer. American Journal of Bioscience and Bioengineering, 4(2), 26-33. https://doi.org/10.11648/j.bio.20160402.12

Copy | Download

ACS Style

Wei Hu. Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer. Am. J. BioSci. Bioeng. 2016, 4(2), 26-33. doi: 10.11648/j.bio.20160402.12

Copy | Download

AMA Style

Wei Hu. Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer. Am J BioSci Bioeng. 2016;4(2):26-33. doi: 10.11648/j.bio.20160402.12

Copy | Download

@article{10.11648/j.bio.20160402.12,
  author = {Wei Hu},
  title = {Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer},
  journal = {American Journal of Bioscience and Bioengineering},
  volume = {4},
  number = {2},
  pages = {26-33},
  doi = {10.11648/j.bio.20160402.12},
  url = {https://doi.org/10.11648/j.bio.20160402.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.bio.20160402.12},
  abstract = {Various gene signatures of chemosensitivity in breast cancer have been identified. When used to build predictors of have chemosensitivity, many of them have their prediction accuracy around 80%. Identifying gene signatures to build high accuracy such predictors is a prerequisite for their clinical tests and applications. To elucidate the importance of each individual gene in a signature is another pressing need before such signature could be tested in clinical settings. In this study, Genetic Algorithms (GAs) and Sparse Logistic Regression (SLR) were employed to identify two signatures. The first had 28 probe sets selected by GA from the top 65 probe sets that were highly overexpressed between pathologic compete response (pCR) and residual disease (RD) and was used to build a SLR predictor of pCR (SLR-28). The second had 86 probe sets (Notch-86) selected by GA from Notch signaling pathway and was used to develop a SLR predictor of pCR (SLR-Notch-86). These two predictors tested on a training set (n=81) and validation set (n=52) had very precise predictions measured by accuracy, specificity, sensitivity, positive predictive value and negative predictive value with their corresponding P value all zero. Furthermore, these two predictors discovered 12 important genes in the 28 probe set signature and 14 important genes in the Notch-86 signature. Our two signatures produced superior performance over a signature in a previous study, demonstrating the potential of GA and SLR in identifying robust gene signatures in chemo response prediction in breast cancer.},
 year = {2016}
}

Copy | Download

TY  - JOUR
T1  - Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer
AU  - Wei Hu
Y1  - 2016/05/04
PY  - 2016
N1  - https://doi.org/10.11648/j.bio.20160402.12
DO  - 10.11648/j.bio.20160402.12
T2  - American Journal of Bioscience and Bioengineering
JF  - American Journal of Bioscience and Bioengineering
JO  - American Journal of Bioscience and Bioengineering
SP  - 26
EP  - 33
PB  - Science Publishing Group
SN  - 2328-5893
UR  - https://doi.org/10.11648/j.bio.20160402.12
AB  - Various gene signatures of chemosensitivity in breast cancer have been identified. When used to build predictors of have chemosensitivity, many of them have their prediction accuracy around 80%. Identifying gene signatures to build high accuracy such predictors is a prerequisite for their clinical tests and applications. To elucidate the importance of each individual gene in a signature is another pressing need before such signature could be tested in clinical settings. In this study, Genetic Algorithms (GAs) and Sparse Logistic Regression (SLR) were employed to identify two signatures. The first had 28 probe sets selected by GA from the top 65 probe sets that were highly overexpressed between pathologic compete response (pCR) and residual disease (RD) and was used to build a SLR predictor of pCR (SLR-28). The second had 86 probe sets (Notch-86) selected by GA from Notch signaling pathway and was used to develop a SLR predictor of pCR (SLR-Notch-86). These two predictors tested on a training set (n=81) and validation set (n=52) had very precise predictions measured by accuracy, specificity, sensitivity, positive predictive value and negative predictive value with their corresponding P value all zero. Furthermore, these two predictors discovered 12 important genes in the 28 probe set signature and 14 important genes in the Notch-86 signature. Our two signatures produced superior performance over a signature in a previous study, demonstrating the potential of GA and SLR in identifying robust gene signatures in chemo response prediction in breast cancer.
VL  - 4
IS  - 2
ER  -

Copy | Download

Author Information

Wei Hu

Department of Computer Science, Houghton College, Houghton, New York, USA

Download PDF

Sections

Plain Text BibTeX RIS

APA Style

Wei Hu. (2016). Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer. American Journal of Bioscience and Bioengineering, 4(2), 26-33. https://doi.org/10.11648/j.bio.20160402.12

Copy | Download

ACS Style

Wei Hu. Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer. Am. J. BioSci. Bioeng. 2016, 4(2), 26-33. doi: 10.11648/j.bio.20160402.12

Copy | Download

AMA Style

Wei Hu. Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer. Am J BioSci Bioeng. 2016;4(2):26-33. doi: 10.11648/j.bio.20160402.12

Copy | Download

@article{10.11648/j.bio.20160402.12,
  author = {Wei Hu},
  title = {Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer},
  journal = {American Journal of Bioscience and Bioengineering},
  volume = {4},
  number = {2},
  pages = {26-33},
  doi = {10.11648/j.bio.20160402.12},
  url = {https://doi.org/10.11648/j.bio.20160402.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.bio.20160402.12},
  abstract = {Various gene signatures of chemosensitivity in breast cancer have been identified. When used to build predictors of have chemosensitivity, many of them have their prediction accuracy around 80%. Identifying gene signatures to build high accuracy such predictors is a prerequisite for their clinical tests and applications. To elucidate the importance of each individual gene in a signature is another pressing need before such signature could be tested in clinical settings. In this study, Genetic Algorithms (GAs) and Sparse Logistic Regression (SLR) were employed to identify two signatures. The first had 28 probe sets selected by GA from the top 65 probe sets that were highly overexpressed between pathologic compete response (pCR) and residual disease (RD) and was used to build a SLR predictor of pCR (SLR-28). The second had 86 probe sets (Notch-86) selected by GA from Notch signaling pathway and was used to develop a SLR predictor of pCR (SLR-Notch-86). These two predictors tested on a training set (n=81) and validation set (n=52) had very precise predictions measured by accuracy, specificity, sensitivity, positive predictive value and negative predictive value with their corresponding P value all zero. Furthermore, these two predictors discovered 12 important genes in the 28 probe set signature and 14 important genes in the Notch-86 signature. Our two signatures produced superior performance over a signature in a previous study, demonstrating the potential of GA and SLR in identifying robust gene signatures in chemo response prediction in breast cancer.},
 year = {2016}
}

Copy | Download

TY  - JOUR
T1  - Using Genetic Algorithms and Sparse Logistic Regression to Find Gene Signatures for Chemosensitivity Prediction in Breast Cancer
AU  - Wei Hu
Y1  - 2016/05/04
PY  - 2016
N1  - https://doi.org/10.11648/j.bio.20160402.12
DO  - 10.11648/j.bio.20160402.12
T2  - American Journal of Bioscience and Bioengineering
JF  - American Journal of Bioscience and Bioengineering
JO  - American Journal of Bioscience and Bioengineering
SP  - 26
EP  - 33
PB  - Science Publishing Group
SN  - 2328-5893
UR  - https://doi.org/10.11648/j.bio.20160402.12
AB  - Various gene signatures of chemosensitivity in breast cancer have been identified. When used to build predictors of have chemosensitivity, many of them have their prediction accuracy around 80%. Identifying gene signatures to build high accuracy such predictors is a prerequisite for their clinical tests and applications. To elucidate the importance of each individual gene in a signature is another pressing need before such signature could be tested in clinical settings. In this study, Genetic Algorithms (GAs) and Sparse Logistic Regression (SLR) were employed to identify two signatures. The first had 28 probe sets selected by GA from the top 65 probe sets that were highly overexpressed between pathologic compete response (pCR) and residual disease (RD) and was used to build a SLR predictor of pCR (SLR-28). The second had 86 probe sets (Notch-86) selected by GA from Notch signaling pathway and was used to develop a SLR predictor of pCR (SLR-Notch-86). These two predictors tested on a training set (n=81) and validation set (n=52) had very precise predictions measured by accuracy, specificity, sensitivity, positive predictive value and negative predictive value with their corresponding P value all zero. Furthermore, these two predictors discovered 12 important genes in the 28 probe set signature and 14 important genes in the Notch-86 signature. Our two signatures produced superior performance over a signature in a previous study, demonstrating the potential of GA and SLR in identifying robust gene signatures in chemo response prediction in breast cancer.
VL  - 4
IS  - 2
ER  -

Copy | Download