Evaluation of methods for modeling transcription factor sequence specificity

Matthew T. Weirauch; Atina Cote; Raquel Norel; Matti Annala; Yue Zhao; Todd R. Riley; Julio Saez-Rodriguez; Thomas Cokelaer; Anastasia Vedenko; Shaheynoor Talukder; Harmen J. Bussemaker; Morris D. Quaid; Martha L. Bulyk; Gustavo Stolovitzky; Timothy R. Hughes; Phaedra Agius; Aaron Arvey; Philipp Bucher; Curtis G. Callan; Cheng Wei Chang; Chien Yu Chen; Yong Syuan Chen; Yu Wei Chu; Jan Grau; Ivo Grosse; Vidhya Jagannathan; Jens Keilwagen; Szymon M. Kiebasa; Justin B. Kinney; Holger Klein; Miron B. Kursa; Harri Lähdesmäki; Kirsti Laurila; Chengwei Lei; Christina Leslie; Chaim Linhart; Anand Murugan; Alena Myšičková; William Stafford Noble; Matti Nykter; Yaron Orenstein; Stefan Posch; Jianhua Ruan; Witold R. Rudnicki; Christoph D. Schmid; Ron Shamir; Wing Kin Sung; Martin Vingron; Zhizhuo Zhang

doi:10.1038/nbt.2486

Evaluation of methods for modeling transcription factor sequence specificity

Matthew T. Weirauch, Atina Cote, Raquel Norel, Matti Annala, Yue Zhao, Todd R. Riley, Julio Saez-Rodriguez, Thomas Cokelaer, Anastasia Vedenko, Shaheynoor Talukder, Harmen J. Bussemaker, Morris D. Quaid, Martha L. Bulyk, Gustavo Stolovitzky, Timothy R. Hughes, Phaedra Agius, Aaron Arvey, Philipp Bucher, Curtis G. Callan, Cheng Wei ChangChien Yu Chen, Yong Syuan Chen, Yu Wei Chu, Jan Grau, Ivo Grosse, Vidhya Jagannathan, Jens Keilwagen, Szymon M. Kiebasa, Justin B. Kinney, Holger Klein, Miron B. Kursa, Harri Lähdesmäki, Kirsti Laurila, Chengwei Lei, Christina Leslie, Chaim Linhart, Anand Murugan, Alena Myšičková, William Stafford Noble, Matti Nykter, Yaron Orenstein, Stefan Posch, Jianhua Ruan, Witold R. Rudnicki, Christoph D. Schmid, Ron Shamir, Wing Kin Sung, Martin Vingron, Zhizhuo Zhang

Research output: Contribution to journal › Article › peer-review

282 Scopus citations

Abstract

Genomic analyses often involve scanning for potential transcription factor (TF) binding sites using models of the sequence specificity of DNA binding proteins. Many approaches have been developed to model and learn a protein's DNA-binding specificity, but these methods have not been systematically compared. Here we applied 26 such approaches to in vitro protein binding microarray data for 66 mouse TFs belonging to various families. For nine TFs, we also scored the resulting motif models on in vivo data, and found that the best in vitro-derived motifs performed similarly to motifs derived from the in vivo data. Our results indicate that simple models based on mononucleotide position weight matrices trained by the best methods perform similarly to more complex models for most TFs examined, but fall short in specific cases (<10% of the TFs examined here). In addition, the best-performing motifs typically have relatively low information content, consistent with widespread degeneracy in eukaryotic TF sequence preferences.

Original language	English
Pages (from-to)	126-134
Number of pages	9
Journal	Nature Biotechnology
Volume	31
Issue number	2
DOIs	https://doi.org/10.1038/nbt.2486
State	Published - Feb 2013
Externally published	Yes

Access to Document

10.1038/nbt.2486

Cite this

Weirauch, M. T., Cote, A., Norel, R., Annala, M., Zhao, Y., Riley, T. R., Saez-Rodriguez, J., Cokelaer, T., Vedenko, A., Talukder, S., Bussemaker, H. J., Quaid, M. D., Bulyk, M. L., Stolovitzky, G., Hughes, T. R., Agius, P., Arvey, A., Bucher, P., Callan, C. G., ... Zhang, Z. (2013). Evaluation of methods for modeling transcription factor sequence specificity. Nature Biotechnology, 31(2), 126-134. https://doi.org/10.1038/nbt.2486

@article{da3496711e864546b42125acbee00f6a,

title = "Evaluation of methods for modeling transcription factor sequence specificity",

abstract = "Genomic analyses often involve scanning for potential transcription factor (TF) binding sites using models of the sequence specificity of DNA binding proteins. Many approaches have been developed to model and learn a protein's DNA-binding specificity, but these methods have not been systematically compared. Here we applied 26 such approaches to in vitro protein binding microarray data for 66 mouse TFs belonging to various families. For nine TFs, we also scored the resulting motif models on in vivo data, and found that the best in vitro-derived motifs performed similarly to motifs derived from the in vivo data. Our results indicate that simple models based on mononucleotide position weight matrices trained by the best methods perform similarly to more complex models for most TFs examined, but fall short in specific cases (<10% of the TFs examined here). In addition, the best-performing motifs typically have relatively low information content, consistent with widespread degeneracy in eukaryotic TF sequence preferences.",

author = "Weirauch, {Matthew T.} and Atina Cote and Raquel Norel and Matti Annala and Yue Zhao and Riley, {Todd R.} and Julio Saez-Rodriguez and Thomas Cokelaer and Anastasia Vedenko and Shaheynoor Talukder and Bussemaker, {Harmen J.} and Quaid, {Morris D.} and Bulyk, {Martha L.} and Gustavo Stolovitzky and Hughes, {Timothy R.} and Phaedra Agius and Aaron Arvey and Philipp Bucher and Callan, {Curtis G.} and Chang, {Cheng Wei} and Chen, {Chien Yu} and Chen, {Yong Syuan} and Chu, {Yu Wei} and Jan Grau and Ivo Grosse and Vidhya Jagannathan and Jens Keilwagen and Kiebasa, {Szymon M.} and Kinney, {Justin B.} and Holger Klein and Kursa, {Miron B.} and Harri L{\"a}hdesm{\"a}ki and Kirsti Laurila and Chengwei Lei and Christina Leslie and Chaim Linhart and Anand Murugan and Alena My{\v s}i{\v c}kov{\'a} and Noble, {William Stafford} and Matti Nykter and Yaron Orenstein and Stefan Posch and Jianhua Ruan and Rudnicki, {Witold R.} and Schmid, {Christoph D.} and Ron Shamir and Sung, {Wing Kin} and Martin Vingron and Zhizhuo Zhang",

note = "Funding Information: were supported by the Academy of Finland (project 260403) and EU ERASysBio ERA-NET. Y.O., C.L. and R.S. were funded by the European Community{\textquoteright}s Seventh Framework Programme under grant agreement no. HEALTH-F4-2009-223575 for the TRIREME project, and by the Israel Science Foundation (grant no. 802/08). Y.O. was supported in part by a fellowship from the Edmond J. Safra Bioinformatics Program at Tel Aviv University. J.G., I.G., S.P. and J.K. were supported by grant XP3624HP/0606T by the Ministry of Culture of Saxony-Anhalt. A.M. was supported by US National Science Foundation (NSF) grant PHY-1022140. C.C. was supported by NSF grant PHY-0957573. J.B.K. was supported by the Simons Center for Quantitative Biology at Cold Spring Harbor Laboratory. Funding Information: We thank H. van Bakel and M. Albu for database assistance, and members of the Hughes laboratory for helpful discussion. M.T.W. was supported by fellowships from the Canadian Institutes of Health Research (CIHR) and the Canadian Institute for Advanced Research (CIFAR) Junior Fellows Genetic Networks Program. This work was supported in part by the Ontario Research Fund and Genome Canada through the Ontario Genomics Institute, and the March of Dimes (T.R.H.). Funding was also provided by Operating Grant MOP-77721 from CIHR to T.R.H. and M.L.B., and grant no. R01 HG003985 from the US National Institutes of Health/National Human Genome Research Institute to M.L.B., as well as US National Institutes of Health grants R01HG003008 and U54CA121852 and a John Simon Guggenheim Foundation Fellowship to H.J.B. M.A., K.L., H.L. and M.L.",

year = "2013",

month = feb,

doi = "10.1038/nbt.2486",

language = "English",

volume = "31",

pages = "126--134",

journal = "Nature Biotechnology",

issn = "1087-0156",

publisher = "Nature Publishing Group",

number = "2",

}

Weirauch, MT, Cote, A, Norel, R, Annala, M, Zhao, Y, Riley, TR, Saez-Rodriguez, J, Cokelaer, T, Vedenko, A, Talukder, S, Bussemaker, HJ, Quaid, MD, Bulyk, ML, Stolovitzky, G, Hughes, TR, Agius, P, Arvey, A, Bucher, P, Callan, CG, Chang, CW, Chen, CY, Chen, YS, Chu, YW, Grau, J, Grosse, I, Jagannathan, V, Keilwagen, J, Kiebasa, SM, Kinney, JB, Klein, H, Kursa, MB, Lähdesmäki, H, Laurila, K, Lei, C, Leslie, C, Linhart, C, Murugan, A, Myšičková, A, Noble, WS, Nykter, M, Orenstein, Y, Posch, S, Ruan, J, Rudnicki, WR, Schmid, CD, Shamir, R, Sung, WK, Vingron, M & Zhang, Z 2013, 'Evaluation of methods for modeling transcription factor sequence specificity', Nature Biotechnology, vol. 31, no. 2, pp. 126-134. https://doi.org/10.1038/nbt.2486

TY - JOUR

T1 - Evaluation of methods for modeling transcription factor sequence specificity

AU - Weirauch, Matthew T.

AU - Cote, Atina

AU - Norel, Raquel

AU - Annala, Matti

AU - Zhao, Yue

AU - Riley, Todd R.

AU - Saez-Rodriguez, Julio

AU - Cokelaer, Thomas

AU - Vedenko, Anastasia

AU - Talukder, Shaheynoor

AU - Bussemaker, Harmen J.

AU - Quaid, Morris D.

AU - Bulyk, Martha L.

AU - Stolovitzky, Gustavo

AU - Hughes, Timothy R.

AU - Agius, Phaedra

AU - Arvey, Aaron

AU - Bucher, Philipp

AU - Callan, Curtis G.

AU - Chang, Cheng Wei

AU - Chen, Chien Yu

AU - Chen, Yong Syuan

AU - Chu, Yu Wei

AU - Grau, Jan

AU - Grosse, Ivo

AU - Jagannathan, Vidhya

AU - Keilwagen, Jens

AU - Kiebasa, Szymon M.

AU - Kinney, Justin B.

AU - Klein, Holger

AU - Kursa, Miron B.

AU - Lähdesmäki, Harri

AU - Laurila, Kirsti

AU - Lei, Chengwei

AU - Leslie, Christina

AU - Linhart, Chaim

AU - Murugan, Anand

AU - Myšičková, Alena

AU - Noble, William Stafford

AU - Nykter, Matti

AU - Orenstein, Yaron

AU - Posch, Stefan

AU - Ruan, Jianhua

AU - Rudnicki, Witold R.

AU - Schmid, Christoph D.

AU - Shamir, Ron

AU - Sung, Wing Kin

AU - Vingron, Martin

AU - Zhang, Zhizhuo

N1 - Funding Information: were supported by the Academy of Finland (project 260403) and EU ERASysBio ERA-NET. Y.O., C.L. and R.S. were funded by the European Community’s Seventh Framework Programme under grant agreement no. HEALTH-F4-2009-223575 for the TRIREME project, and by the Israel Science Foundation (grant no. 802/08). Y.O. was supported in part by a fellowship from the Edmond J. Safra Bioinformatics Program at Tel Aviv University. J.G., I.G., S.P. and J.K. were supported by grant XP3624HP/0606T by the Ministry of Culture of Saxony-Anhalt. A.M. was supported by US National Science Foundation (NSF) grant PHY-1022140. C.C. was supported by NSF grant PHY-0957573. J.B.K. was supported by the Simons Center for Quantitative Biology at Cold Spring Harbor Laboratory. Funding Information: We thank H. van Bakel and M. Albu for database assistance, and members of the Hughes laboratory for helpful discussion. M.T.W. was supported by fellowships from the Canadian Institutes of Health Research (CIHR) and the Canadian Institute for Advanced Research (CIFAR) Junior Fellows Genetic Networks Program. This work was supported in part by the Ontario Research Fund and Genome Canada through the Ontario Genomics Institute, and the March of Dimes (T.R.H.). Funding was also provided by Operating Grant MOP-77721 from CIHR to T.R.H. and M.L.B., and grant no. R01 HG003985 from the US National Institutes of Health/National Human Genome Research Institute to M.L.B., as well as US National Institutes of Health grants R01HG003008 and U54CA121852 and a John Simon Guggenheim Foundation Fellowship to H.J.B. M.A., K.L., H.L. and M.L.

PY - 2013/2

Y1 - 2013/2

N2 - Genomic analyses often involve scanning for potential transcription factor (TF) binding sites using models of the sequence specificity of DNA binding proteins. Many approaches have been developed to model and learn a protein's DNA-binding specificity, but these methods have not been systematically compared. Here we applied 26 such approaches to in vitro protein binding microarray data for 66 mouse TFs belonging to various families. For nine TFs, we also scored the resulting motif models on in vivo data, and found that the best in vitro-derived motifs performed similarly to motifs derived from the in vivo data. Our results indicate that simple models based on mononucleotide position weight matrices trained by the best methods perform similarly to more complex models for most TFs examined, but fall short in specific cases (<10% of the TFs examined here). In addition, the best-performing motifs typically have relatively low information content, consistent with widespread degeneracy in eukaryotic TF sequence preferences.

AB - Genomic analyses often involve scanning for potential transcription factor (TF) binding sites using models of the sequence specificity of DNA binding proteins. Many approaches have been developed to model and learn a protein's DNA-binding specificity, but these methods have not been systematically compared. Here we applied 26 such approaches to in vitro protein binding microarray data for 66 mouse TFs belonging to various families. For nine TFs, we also scored the resulting motif models on in vivo data, and found that the best in vitro-derived motifs performed similarly to motifs derived from the in vivo data. Our results indicate that simple models based on mononucleotide position weight matrices trained by the best methods perform similarly to more complex models for most TFs examined, but fall short in specific cases (<10% of the TFs examined here). In addition, the best-performing motifs typically have relatively low information content, consistent with widespread degeneracy in eukaryotic TF sequence preferences.

UR - http://www.scopus.com/inward/record.url?scp=84873594686&partnerID=8YFLogxK

U2 - 10.1038/nbt.2486

DO - 10.1038/nbt.2486

M3 - Article

C2 - 23354101

AN - SCOPUS:84873594686

SN - 1087-0156

VL - 31

SP - 126

EP - 134

JO - Nature Biotechnology

JF - Nature Biotechnology

IS - 2

ER -

Evaluation of methods for modeling transcription factor sequence specificity

Abstract

Access to Document

Fingerprint

Cite this