Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics

Mi Yang; Francesca Petralia; Zhi Li; Hongyang Li; Weiping Ma; Xiaoyu Song; Sunkyu Kim; Heewon Lee; Han Yu; Bora Lee; Seohui Bae; Eunji Heo; Jan Kaczmarczyk; Piotr Stępniak; Michał Warchoł; Thomas Yu; Anna P. Calinawan; Paul C. Boutros; Samuel H. Payne; Boris Reva; Tunde Aderinwale; Ebrahim Afyounian; Piyush Agrawal; Mehreen Ali; Alicia Amadoz; Francisco Azuaje; John Bachman; Sherry Bhalla; José Carbonell-Caballero; Priyanka Chakraborty; Kumardeep Chaudhary; Yonghwa Choi; Yoonjung Choi; Cankut Çubuk; Sandeep Kumar Dhanda; Joaquín Dopazo; Laura L. Elo; Ábel Fóthi; Olivier Gevaert; Kirsi Granberg; Russell Greiner; Marta R. Hidalgo; Vivek Jayaswal; Hwisang Jeon; Minji Jeon; Sunil V. Kalmady; Yasuhiro Kambara; Sang Yoon Kim; Arnaud Muller; Petr V. Nazarov; NCI-CPTAC-DREAM Consortium; Léon-Charles Tranchevent

doi:10.1016/j.cels.2020.06.013

Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics

Mi Yang, Francesca Petralia, Zhi Li, Hongyang Li, Weiping Ma, Xiaoyu Song, Sunkyu Kim, Heewon Lee, Han Yu, Bora Lee, Seohui Bae, Eunji Heo, Jan Kaczmarczyk, Piotr Stępniak, Michał Warchoł, Thomas Yu, Anna P. Calinawan, Paul C. Boutros, Samuel H. Payne, Boris RevaTunde Aderinwale, Ebrahim Afyounian, Piyush Agrawal, Mehreen Ali, Alicia Amadoz, Francisco Azuaje, John Bachman, Sherry Bhalla, José Carbonell-Caballero, Priyanka Chakraborty, Kumardeep Chaudhary, Yonghwa Choi, Yoonjung Choi, Cankut Çubuk, Sandeep Kumar Dhanda, Joaquín Dopazo, Laura L. Elo, Ábel Fóthi, Olivier Gevaert, Kirsi Granberg, Russell Greiner, Marta R. Hidalgo, Vivek Jayaswal, Hwisang Jeon, Minji Jeon, Sunil V. Kalmady, Yasuhiro Kambara, Sang Yoon Kim, Arnaud Muller, Petr V. Nazarov, NCI-CPTAC-DREAM Consortium, Léon-Charles Tranchevent

Multiomics Data Science

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

14 Citations (Scopus)

Abstract

A major manifestation of cancer is the alteration of protein measurements. However, proteins are harder and more expensive to measure than genes and transcripts. To address this problem, we crowdsourced it via the NCI-CPTAC DREAM proteogenomics challenge. We provided participants data to build models to predict protein and phosphorylation levels from genomic and transcriptomic data in cancer patients. We then asked participants to use such models to predict unseen (phospho)protein data from given genomic and transcriptomic data in other patients. This experiment allowed us to assess the predictive performance of the proposed methods in an unbiased and “double-blinded” manner. We found that ensemble methods perform better, and we identified which proteins and biological processes are easier or harder to predict. In general, performance was limited, suggesting that (phospho)proteomic cannot be replaced, at least yet, by genomic and transcriptomic profiling.

Original language	English
Pages (from-to)	186-195.e9
Journal	Cell Systems
Volume	11
Issue number	2
DOIs	https://doi.org/10.1016/j.cels.2020.06.013
Publication status	Published - 26 Aug 2020

Keywords

cancer
crowdsourcing
genomics
machine learning
protein regulation
proteogenomics
proteomics

Access to Document

10.1016/j.cels.2020.06.013

Cite this

Yang, M., Petralia, F., Li, Z., Li, H., Ma, W., Song, X., Kim, S., Lee, H., Yu, H., Lee, B., Bae, S., Heo, E., Kaczmarczyk, J., Stępniak, P., Warchoł, M., Yu, T., Calinawan, A. P., Boutros, P. C., Payne, S. H., ... Tranchevent , L.-C. (2020). Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics. Cell Systems, 11(2), 186-195.e9. https://doi.org/10.1016/j.cels.2020.06.013

@article{bdc3f93bfc4a4f40b97bded64c5673d8,

title = "Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics",

abstract = "A major manifestation of cancer is the alteration of protein measurements. However, proteins are harder and more expensive to measure than genes and transcripts. To address this problem, we crowdsourced it via the NCI-CPTAC DREAM proteogenomics challenge. We provided participants data to build models to predict protein and phosphorylation levels from genomic and transcriptomic data in cancer patients. We then asked participants to use such models to predict unseen (phospho)protein data from given genomic and transcriptomic data in other patients. This experiment allowed us to assess the predictive performance of the proposed methods in an unbiased and “double-blinded” manner. We found that ensemble methods perform better, and we identified which proteins and biological processes are easier or harder to predict. In general, performance was limited, suggesting that (phospho)proteomic cannot be replaced, at least yet, by genomic and transcriptomic profiling.",

keywords = "cancer, crowdsourcing, genomics, machine learning, protein regulation, proteogenomics, proteomics",

author = "Mi Yang and Francesca Petralia and Zhi Li and Hongyang Li and Weiping Ma and Xiaoyu Song and Sunkyu Kim and Heewon Lee and Han Yu and Bora Lee and Seohui Bae and Eunji Heo and Jan Kaczmarczyk and Piotr St{\c e}pniak and Micha{\l} Warcho{\l} and Thomas Yu and Calinawan, {Anna P.} and Boutros, {Paul C.} and Payne, {Samuel H.} and Boris Reva and Tunde Aderinwale and Ebrahim Afyounian and Piyush Agrawal and Mehreen Ali and Alicia Amadoz and Francisco Azuaje and John Bachman and Sherry Bhalla and Jos{\'e} Carbonell-Caballero and Priyanka Chakraborty and Kumardeep Chaudhary and Yonghwa Choi and Yoonjung Choi and Cankut {\c C}ubuk and Dhanda, {Sandeep Kumar} and Joaqu{\'i}n Dopazo and Elo, {Laura L.} and {\'A}bel F{\'o}thi and Olivier Gevaert and Kirsi Granberg and Russell Greiner and Hidalgo, {Marta R.} and Vivek Jayaswal and Hwisang Jeon and Minji Jeon and Kalmady, {Sunil V.} and Yasuhiro Kambara and Kim, {Sang Yoon} and Arnaud Muller and Nazarov, {Petr V.} and {NCI-CPTAC-DREAM Consortium} and L{\'e}on-Charles Tranchevent",

note = "Funding Information: We would like to thank Nicolas Palacio and Emanuel Gon?alves for their suggestions and ideas in improving the manuscript. We would also like to thank the National Cancer Institute's Clinical Proteomic Tumor Analysis Consortium (CPTAC), a comprehensive and coordinated effort to accelerate the understanding of the molecular basis of cancer through the application of proteogenomics, for providing the data sets used in this challenge and making it freely available to the public. The challenge was supported financially by the CPTAC. This research work is supported by the National Institutes for Health (NIH), National Cancer Institute (NCI) CPTAC awards U24 CA210972 and U24 CA210993, and a contract 16X230 from Leidos Biomedical Research. M.Y. F.P. Z.L. W.M. D.F. P.C.B. E.B. H.R. and J.S.-R. organized the challenge. M.Y. F.P. X.S. H.L. Y.G. and Z.L. performed the post challenge analysis and wrote the manuscript. M.Y. developed the Shiny application. M.Y. and F.P. tested baseline solutions for the challenge questions. H.L. and Y.G. developed the best-performing algorithms of the challenge. F.P. organized the collaborative round. S.K. H.L. H.Y. B.L. S.B. E.H. J.K. P.S. M.W. Y.G. and J.K. participated in the collaborative round and the writing of their methods. Z.L. processed the data used in the challenge. M.Y. assisted in processing the challenge data. T.Y. implemented the challenge infrastructure. P.C.B. assisted in challenge design. P.C.B. S.H.P. H.R. Y.G. and G.S. commented on the manuscript and provided meaningful feedback. P.W. provided meaningful feedback at all stages. H.R. initiated the challenge. D.F. and J.S.-R. supervised the project. NCI-CPTAC-DREAM consortium participated in the challenge and submitted their predictions. The authors declare no competing interests. Funding Information: We would like to thank Nicolas Palacio and Emanuel Gon{\c c}alves for their suggestions and ideas in improving the manuscript. We would also like to thank the National Cancer Institute{\textquoteright}s Clinical Proteomic Tumor Analysis Consortium (CPTAC), a comprehensive and coordinated effort to accelerate the understanding of the molecular basis of cancer through the application of proteogenomics, for providing the data sets used in this challenge and making it freely available to the public. The challenge was supported financially by the CPTAC. This research work is supported by the National Institutes for Health (NIH), National Cancer Institute (NCI) CPTAC awards U24 CA210972 and U24 CA210993 , and a contract 16X230 from Leidos Biomedical Research . Publisher Copyright: {\textcopyright} 2020 The Authors",

year = "2020",

month = aug,

day = "26",

doi = "10.1016/j.cels.2020.06.013",

language = "English",

volume = "11",

pages = "186--195.e9",

journal = "Cell Systems",

issn = "2405-4712",

publisher = "Cell Press",

number = "2",

}

Yang, M, Petralia, F, Li, Z, Li, H, Ma, W, Song, X, Kim, S, Lee, H, Yu, H, Lee, B, Bae, S, Heo, E, Kaczmarczyk, J, Stępniak, P, Warchoł, M, Yu, T, Calinawan, AP, Boutros, PC, Payne, SH, Reva, B, Aderinwale, T, Afyounian, E, Agrawal, P, Ali, M, Amadoz, A, Azuaje, F, Bachman, J, Bhalla, S, Carbonell-Caballero, J, Chakraborty, P, Chaudhary, K, Choi, Y, Choi, Y, Çubuk, C, Dhanda, SK, Dopazo, J, Elo, LL, Fóthi, Á, Gevaert, O, Granberg, K, Greiner, R, Hidalgo, MR, Jayaswal, V, Jeon, H, Jeon, M, Kalmady, SV, Kambara, Y, Kim, SY, Muller, A, Nazarov, PV, NCI-CPTAC-DREAM Consortium & Tranchevent , L-C 2020, 'Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics', Cell Systems, vol. 11, no. 2, pp. 186-195.e9. https://doi.org/10.1016/j.cels.2020.06.013

TY - JOUR

T1 - Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics

AU - Yang, Mi

AU - Petralia, Francesca

AU - Li, Zhi

AU - Li, Hongyang

AU - Ma, Weiping

AU - Song, Xiaoyu

AU - Kim, Sunkyu

AU - Lee, Heewon

AU - Yu, Han

AU - Lee, Bora

AU - Bae, Seohui

AU - Heo, Eunji

AU - Kaczmarczyk, Jan

AU - Stępniak, Piotr

AU - Warchoł, Michał

AU - Yu, Thomas

AU - Calinawan, Anna P.

AU - Boutros, Paul C.

AU - Payne, Samuel H.

AU - Reva, Boris

AU - Aderinwale, Tunde

AU - Afyounian, Ebrahim

AU - Agrawal, Piyush

AU - Ali, Mehreen

AU - Amadoz, Alicia

AU - Azuaje, Francisco

AU - Bachman, John

AU - Bhalla, Sherry

AU - Carbonell-Caballero, José

AU - Chakraborty, Priyanka

AU - Chaudhary, Kumardeep

AU - Choi, Yonghwa

AU - Choi, Yoonjung

AU - Çubuk, Cankut

AU - Dhanda, Sandeep Kumar

AU - Dopazo, Joaquín

AU - Elo, Laura L.

AU - Fóthi, Ábel

AU - Gevaert, Olivier

AU - Granberg, Kirsi

AU - Greiner, Russell

AU - Hidalgo, Marta R.

AU - Jayaswal, Vivek

AU - Jeon, Hwisang

AU - Jeon, Minji

AU - Kalmady, Sunil V.

AU - Kambara, Yasuhiro

AU - Kim, Sang Yoon

AU - Muller, Arnaud

AU - Nazarov, Petr V.

AU - NCI-CPTAC-DREAM Consortium

AU - Tranchevent , Léon-Charles

N1 - Funding Information: We would like to thank Nicolas Palacio and Emanuel Gon?alves for their suggestions and ideas in improving the manuscript. We would also like to thank the National Cancer Institute's Clinical Proteomic Tumor Analysis Consortium (CPTAC), a comprehensive and coordinated effort to accelerate the understanding of the molecular basis of cancer through the application of proteogenomics, for providing the data sets used in this challenge and making it freely available to the public. The challenge was supported financially by the CPTAC. This research work is supported by the National Institutes for Health (NIH), National Cancer Institute (NCI) CPTAC awards U24 CA210972 and U24 CA210993, and a contract 16X230 from Leidos Biomedical Research. M.Y. F.P. Z.L. W.M. D.F. P.C.B. E.B. H.R. and J.S.-R. organized the challenge. M.Y. F.P. X.S. H.L. Y.G. and Z.L. performed the post challenge analysis and wrote the manuscript. M.Y. developed the Shiny application. M.Y. and F.P. tested baseline solutions for the challenge questions. H.L. and Y.G. developed the best-performing algorithms of the challenge. F.P. organized the collaborative round. S.K. H.L. H.Y. B.L. S.B. E.H. J.K. P.S. M.W. Y.G. and J.K. participated in the collaborative round and the writing of their methods. Z.L. processed the data used in the challenge. M.Y. assisted in processing the challenge data. T.Y. implemented the challenge infrastructure. P.C.B. assisted in challenge design. P.C.B. S.H.P. H.R. Y.G. and G.S. commented on the manuscript and provided meaningful feedback. P.W. provided meaningful feedback at all stages. H.R. initiated the challenge. D.F. and J.S.-R. supervised the project. NCI-CPTAC-DREAM consortium participated in the challenge and submitted their predictions. The authors declare no competing interests. Funding Information: We would like to thank Nicolas Palacio and Emanuel Gonçalves for their suggestions and ideas in improving the manuscript. We would also like to thank the National Cancer Institute’s Clinical Proteomic Tumor Analysis Consortium (CPTAC), a comprehensive and coordinated effort to accelerate the understanding of the molecular basis of cancer through the application of proteogenomics, for providing the data sets used in this challenge and making it freely available to the public. The challenge was supported financially by the CPTAC. This research work is supported by the National Institutes for Health (NIH), National Cancer Institute (NCI) CPTAC awards U24 CA210972 and U24 CA210993 , and a contract 16X230 from Leidos Biomedical Research . Publisher Copyright: © 2020 The Authors

PY - 2020/8/26

Y1 - 2020/8/26

N2 - A major manifestation of cancer is the alteration of protein measurements. However, proteins are harder and more expensive to measure than genes and transcripts. To address this problem, we crowdsourced it via the NCI-CPTAC DREAM proteogenomics challenge. We provided participants data to build models to predict protein and phosphorylation levels from genomic and transcriptomic data in cancer patients. We then asked participants to use such models to predict unseen (phospho)protein data from given genomic and transcriptomic data in other patients. This experiment allowed us to assess the predictive performance of the proposed methods in an unbiased and “double-blinded” manner. We found that ensemble methods perform better, and we identified which proteins and biological processes are easier or harder to predict. In general, performance was limited, suggesting that (phospho)proteomic cannot be replaced, at least yet, by genomic and transcriptomic profiling.

AB - A major manifestation of cancer is the alteration of protein measurements. However, proteins are harder and more expensive to measure than genes and transcripts. To address this problem, we crowdsourced it via the NCI-CPTAC DREAM proteogenomics challenge. We provided participants data to build models to predict protein and phosphorylation levels from genomic and transcriptomic data in cancer patients. We then asked participants to use such models to predict unseen (phospho)protein data from given genomic and transcriptomic data in other patients. This experiment allowed us to assess the predictive performance of the proposed methods in an unbiased and “double-blinded” manner. We found that ensemble methods perform better, and we identified which proteins and biological processes are easier or harder to predict. In general, performance was limited, suggesting that (phospho)proteomic cannot be replaced, at least yet, by genomic and transcriptomic profiling.

KW - cancer

KW - crowdsourcing

KW - genomics

KW - machine learning

KW - protein regulation

KW - proteogenomics

KW - proteomics

UR - http://www.scopus.com/inward/record.url?scp=85089295261&partnerID=8YFLogxK

UR - https://www.ncbi.nlm.nih.gov/pubmed/32710834

U2 - 10.1016/j.cels.2020.06.013

DO - 10.1016/j.cels.2020.06.013

M3 - Article

C2 - 32710834

AN - SCOPUS:85089295261

SN - 2405-4712

VL - 11

SP - 186-195.e9

JO - Cell Systems

JF - Cell Systems

IS - 2

ER -

Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this