A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns

PCAWG Tumor Subtypes and Clinical Translation Working Group; PCAWG Consortium; Wei Jiao; Gurnit Atwal; Paz Polak; Rosa Karlic; Edwin Cuppen; Fatima Al-Shahrour; Gurnit Atwal; Peter J. Bailey; Andrew V. Biankin; Paul C. Boutros; Peter J. Campbell; David K. Chang; Susanna L. Cooke; Vikram Deshpande; Bishoy M. Faltas; William C. Faquin; Levi Garraway; Gad Getz; Sean M. Grimmond; Syed Haider; Katherine A. Hoadley; Wei Jiao; Vera B. Kaiser; Rosa Karlic; Mamoru Kato; Kirsten Kübler; Alexander J. Lazar; Constance H. Li; David N. Louis; Adam Margolin; Sancha Martin; Hardeep K. Nahal-Bose; G. Petur Nielsen; Serena Nik-Zainal; Larsson Omberg; Christine P’ng; Marc D. Perry; Paz Polak; Esther Rheinbay; Mark A. Rubin; Colin A. Semple; Dennis C. Sgroi; Tatsuhiro Shibata; Reiner Siebert; Jaclyn Smith; Lincoln D. Stein; Miranda D. Stobbe; Ren X. Sun; Kevin Thai; Derek W. Wright; Chin Lee Wu; Ke Yuan; Junjun Zhang; Alexandra Danyi; Jeroen de Ridder; Carla van Herpen; Martijn P. Lolkema; Neeltje Steeghs; Gad Getz; Quaid D. Morris; Lauri A. Aaltonen; Federico Abascal; Adam Abeshouse; Hiroyuki Aburatani; David J. Adams; Nishant Agrawal; Keun Soo Ahn; Sung Min Ahn; Hiroshi Aikata; Rehan Akbani; Kadir C. Akdemir; Hikmat Al-Ahmadie; Sultan T. Al-Sedairy; Fatima Al-Shahrour; Malik Alawi; Monique Albert; Kenneth Aldape; Ludmil B. Alexandrov; Adrian Ally; Kathryn Alsop; Eva G. Alvarez; Fernanda Amary; Samirkumar B. Amin; Brice Aminou; Ole Ammerpohl; Matthew J. Anderson; Yeng Ang; Davide Antonello; Pavana Anur; Samuel Aparicio; Elizabeth L. Appelbaum; Yasuhito Arai; Axel Aretz; Koji Arihiro; Shun ichi Ariizumi; Joshua Armenia; Laurent Arnould; Sylvia Asa; Yassen Assenov; Gurnit Atwal; Sietse Aukema; J. Todd Auman; Miriam R.R. Aure; Philip Awadalla; Marta Aymerich; Gary D. Bader; Adrian Baez-Ortega; Matthew H. Bailey; Peter J. Bailey; Miruna Balasundaram; Saianand Balu; Pratiti Bandopadhayay; Rosamonde E. Banks; Stefano Barbi; Andrew P. Barbour; Jonathan Barenboim; Jill Barnholtz-Sloan; Hugh Barr; Elisabet Barrera; John Bartlett; Javier Bartolome; Claudio Bassi; Oliver F. Bathe; Daniel Baumhoer; Prashant Bavi; Stephen B. Baylin; Wojciech Bazant; Duncan Beardsmore; Timothy A. Beck; Sam Behjati; Andreas Behren; Beifang Niu; Cindy Bell; Sergi Beltran; Christopher Benz; Andrew Berchuck; Anke K. Bergmann; Erik N. Bergstrom; Benjamin P. Berman; Daniel M. Berney; Stephan H. Bernhart; Rameen Beroukhim; Mario Berrios; Samantha Bersani; Johanna Bertl; Miguel Betancourt; Vinayak Bhandari; Shriram G. Bhosle; Andrew V. Biankin; Matthias Bieg; Darell Bigner; Hans Binder; Ewan Birney; Michael Birrer; Nidhan K. Biswas; Bodil Bjerkehagen; Tom Bodenheimer; Lori Boice; Giada Bonizzato; Johann S. De Bono; Arnoud Boot; Moiz S. Bootwalla; Ake Borg; Arndt Borkhardt; Keith A. Boroevich; Ivan Borozan; Christoph Borst; Marcus Bosenberg; Mattia Bosio; Jacqueline Boultwood; Guillaume Bourque; Paul C. Boutros; G. Steven Bova; David T. Bowen; Reanne Bowlby; David D.L. Bowtell; Sandrine Boyault; Rich Boyce; Jeffrey Boyd; Alvis Brazma; Paul Brennan; Daniel S. Brewer; Arie B. Brinkman; Robert G. Bristow; Russell R. Broaddus; Jane E. Brock; Malcolm Brock; Annegien Broeks; Angela N. Brooks; Denise Brooks; Benedikt Brors; Li Ding; Lucinda A. Fulton; Robert S. Fulton; Ramaswamy Govindan; Reyka Jayasinghe; Tim Ley; Christopher A. Miller; David Mutch; Michael C. Wendl

doi:10.1038/s41467-019-13825-8

A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns

PCAWG Tumor Subtypes and Clinical Translation Working Group, PCAWG Consortium, Wei Jiao, Gurnit Atwal, Paz Polak, Rosa Karlic, Edwin Cuppen, Fatima Al-Shahrour, Gurnit Atwal, Peter J. Bailey, Andrew V. Biankin, Paul C. Boutros, Peter J. Campbell, David K. Chang, Susanna L. Cooke, Vikram Deshpande, Bishoy M. Faltas, William C. Faquin, Levi Garraway, Gad GetzSean M. Grimmond, Syed Haider, Katherine A. Hoadley, Wei Jiao, Vera B. Kaiser, Rosa Karlic, Mamoru Kato, Kirsten Kübler, Alexander J. Lazar, Constance H. Li, David N. Louis, Adam Margolin, Sancha Martin, Hardeep K. Nahal-Bose, G. Petur Nielsen, Serena Nik-Zainal, Larsson Omberg, Christine P’ng, Marc D. Perry, Paz Polak, Esther Rheinbay, Mark A. Rubin, Colin A. Semple, Dennis C. Sgroi, Tatsuhiro Shibata, Reiner Siebert, Jaclyn Smith, Lincoln D. Stein, Miranda D. Stobbe, Ren X. Sun, Kevin Thai, Derek W. Wright, Chin Lee Wu, Ke Yuan, Junjun Zhang, Alexandra Danyi, Jeroen de Ridder, Carla van Herpen, Martijn P. Lolkema, Neeltje Steeghs, Gad Getz, Quaid D. Morris, Lauri A. Aaltonen, Federico Abascal, Adam Abeshouse, Hiroyuki Aburatani, David J. Adams, Nishant Agrawal, Keun Soo Ahn, Sung Min Ahn, Hiroshi Aikata, Rehan Akbani, Kadir C. Akdemir, Hikmat Al-Ahmadie, Sultan T. Al-Sedairy, Fatima Al-Shahrour, Malik Alawi, Monique Albert, Kenneth Aldape, Ludmil B. Alexandrov, Adrian Ally, Kathryn Alsop, Eva G. Alvarez, Fernanda Amary, Samirkumar B. Amin, Brice Aminou, Ole Ammerpohl, Matthew J. Anderson, Yeng Ang, Davide Antonello, Pavana Anur, Samuel Aparicio, Elizabeth L. Appelbaum, Yasuhito Arai, Axel Aretz, Koji Arihiro, Shun ichi Ariizumi, Joshua Armenia, Laurent Arnould, Sylvia Asa, Yassen Assenov, Gurnit Atwal, Sietse Aukema, J. Todd Auman, Miriam R.R. Aure, Philip Awadalla, Marta Aymerich, Gary D. Bader, Adrian Baez-Ortega, Matthew H. Bailey, Peter J. Bailey, Miruna Balasundaram, Saianand Balu, Pratiti Bandopadhayay, Rosamonde E. Banks, Stefano Barbi, Andrew P. Barbour, Jonathan Barenboim, Jill Barnholtz-Sloan, Hugh Barr, Elisabet Barrera, John Bartlett, Javier Bartolome, Claudio Bassi, Oliver F. Bathe, Daniel Baumhoer, Prashant Bavi, Stephen B. Baylin, Wojciech Bazant, Duncan Beardsmore, Timothy A. Beck, Sam Behjati, Andreas Behren, Beifang Niu, Cindy Bell, Sergi Beltran, Christopher Benz, Andrew Berchuck, Anke K. Bergmann, Erik N. Bergstrom, Benjamin P. Berman, Daniel M. Berney, Stephan H. Bernhart, Rameen Beroukhim, Mario Berrios, Samantha Bersani, Johanna Bertl, Miguel Betancourt, Vinayak Bhandari, Shriram G. Bhosle, Andrew V. Biankin, Matthias Bieg, Darell Bigner, Hans Binder, Ewan Birney, Michael Birrer, Nidhan K. Biswas, Bodil Bjerkehagen, Tom Bodenheimer, Lori Boice, Giada Bonizzato, Johann S. De Bono, Arnoud Boot, Moiz S. Bootwalla, Ake Borg, Arndt Borkhardt, Keith A. Boroevich, Ivan Borozan, Christoph Borst, Marcus Bosenberg, Mattia Bosio, Jacqueline Boultwood, Guillaume Bourque, Paul C. Boutros, G. Steven Bova, David T. Bowen, Reanne Bowlby, David D.L. Bowtell, Sandrine Boyault, Rich Boyce, Jeffrey Boyd, Alvis Brazma, Paul Brennan, Daniel S. Brewer, Arie B. Brinkman, Robert G. Bristow, Russell R. Broaddus, Jane E. Brock, Malcolm Brock, Annegien Broeks, Angela N. Brooks, Denise Brooks, Benedikt Brors, Li Ding, Lucinda A. Fulton, Robert S. Fulton, Ramaswamy Govindan, Reyka Jayasinghe, Tim Ley, Christopher A. Miller, David Mutch, Michael C. Wendl

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

134 Scopus citations

Abstract

In cancer, the primary tumour’s organ of origin and histopathology are the strongest determinants of its clinical behaviour, but in 3% of cases a patient presents with a metastatic tumour and no obvious primary. Here,as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, we train a deep learning classifier to predict cancer type based on patterns of somatic passenger mutations detected in whole genome sequencing (WGS) of 2606 tumours representing 24 common cancer types produced by the PCAWG Consortium. Our classifier achieves an accuracy of 91% on held-out tumor samples and 88% and 83% respectively on independent primary and metastatic samples, roughly double the accuracy of trained pathologists when presented with a metastatic tumour without knowledge of the primary. Surprisingly, adding information on driver mutations reduced accuracy. Our results have clinical applicability, underscore how patterns of somatic passenger mutations encode the state of the cell of origin, and can inform future strategies to detect the source of circulating tumour DNA.

Original language	English
Article number	728
Journal	Nature communications
Volume	11
Issue number	1
DOIs	https://doi.org/10.1038/s41467-019-13825-8
State	Published - Dec 1 2020

Access to Document

10.1038/s41467-019-13825-8

Cite this

PCAWG Tumor Subtypes and Clinical Translation Working Group, PCAWG Consortium, Jiao, W., Atwal, G., Polak, P., Karlic, R., Cuppen, E., Al-Shahrour, F., Atwal, G., Bailey, P. J., Biankin, A. V., Boutros, P. C., Campbell, P. J., Chang, D. K., Cooke, S. L., Deshpande, V., Faltas, B. M., Faquin, W. C., Garraway, L., ... Wendl, M. C. (2020). A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns. Nature communications, 11(1), Article 728. https://doi.org/10.1038/s41467-019-13825-8

@article{a989ffa39ad9490fab5cf04b5731e739,

title = "A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns",

abstract = "In cancer, the primary tumour{\textquoteright}s organ of origin and histopathology are the strongest determinants of its clinical behaviour, but in 3% of cases a patient presents with a metastatic tumour and no obvious primary. Here,as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, we train a deep learning classifier to predict cancer type based on patterns of somatic passenger mutations detected in whole genome sequencing (WGS) of 2606 tumours representing 24 common cancer types produced by the PCAWG Consortium. Our classifier achieves an accuracy of 91% on held-out tumor samples and 88% and 83% respectively on independent primary and metastatic samples, roughly double the accuracy of trained pathologists when presented with a metastatic tumour without knowledge of the primary. Surprisingly, adding information on driver mutations reduced accuracy. Our results have clinical applicability, underscore how patterns of somatic passenger mutations encode the state of the cell of origin, and can inform future strategies to detect the source of circulating tumour DNA.",

author = "{PCAWG Tumor Subtypes and Clinical Translation Working Group} and {PCAWG Consortium} and Wei Jiao and Gurnit Atwal and Paz Polak and Rosa Karlic and Edwin Cuppen and Fatima Al-Shahrour and Gurnit Atwal and Bailey, {Peter J.} and Biankin, {Andrew V.} and Boutros, {Paul C.} and Campbell, {Peter J.} and Chang, {David K.} and Cooke, {Susanna L.} and Vikram Deshpande and Faltas, {Bishoy M.} and Faquin, {William C.} and Levi Garraway and Gad Getz and Grimmond, {Sean M.} and Syed Haider and Hoadley, {Katherine A.} and Wei Jiao and Kaiser, {Vera B.} and Rosa Karlic and Mamoru Kato and Kirsten K{\"u}bler and Lazar, {Alexander J.} and Li, {Constance H.} and Louis, {David N.} and Adam Margolin and Sancha Martin and Nahal-Bose, {Hardeep K.} and Nielsen, {G. Petur} and Serena Nik-Zainal and Larsson Omberg and Christine P{\textquoteright}ng and Perry, {Marc D.} and Paz Polak and Esther Rheinbay and Rubin, {Mark A.} and Semple, {Colin A.} and Sgroi, {Dennis C.} and Tatsuhiro Shibata and Reiner Siebert and Jaclyn Smith and Stein, {Lincoln D.} and Stobbe, {Miranda D.} and Sun, {Ren X.} and Kevin Thai and Wright, {Derek W.} and Wu, {Chin Lee} and Ke Yuan and Junjun Zhang and Alexandra Danyi and {de Ridder}, Jeroen and {van Herpen}, Carla and Lolkema, {Martijn P.} and Neeltje Steeghs and Gad Getz and Morris, {Quaid D.} and Aaltonen, {Lauri A.} and Federico Abascal and Adam Abeshouse and Hiroyuki Aburatani and Adams, {David J.} and Nishant Agrawal and Ahn, {Keun Soo} and Ahn, {Sung Min} and Hiroshi Aikata and Rehan Akbani and Akdemir, {Kadir C.} and Hikmat Al-Ahmadie and Al-Sedairy, {Sultan T.} and Fatima Al-Shahrour and Malik Alawi and Monique Albert and Kenneth Aldape and Alexandrov, {Ludmil B.} and Adrian Ally and Kathryn Alsop and Alvarez, {Eva G.} and Fernanda Amary and Amin, {Samirkumar B.} and Brice Aminou and Ole Ammerpohl and Anderson, {Matthew J.} and Yeng Ang and Davide Antonello and Pavana Anur and Samuel Aparicio and Appelbaum, {Elizabeth L.} and Yasuhito Arai and Axel Aretz and Koji Arihiro and Ariizumi, {Shun ichi} and Joshua Armenia and Laurent Arnould and Sylvia Asa and Yassen Assenov and Gurnit Atwal and Sietse Aukema and Auman, {J. Todd} and Aure, {Miriam R.R.} and Philip Awadalla and Marta Aymerich and Bader, {Gary D.} and Adrian Baez-Ortega and Bailey, {Matthew H.} and Bailey, {Peter J.} and Miruna Balasundaram and Saianand Balu and Pratiti Bandopadhayay and Banks, {Rosamonde E.} and Stefano Barbi and Barbour, {Andrew P.} and Jonathan Barenboim and Jill Barnholtz-Sloan and Hugh Barr and Elisabet Barrera and John Bartlett and Javier Bartolome and Claudio Bassi and Bathe, {Oliver F.} and Daniel Baumhoer and Prashant Bavi and Baylin, {Stephen B.} and Wojciech Bazant and Duncan Beardsmore and Beck, {Timothy A.} and Sam Behjati and Andreas Behren and Beifang Niu and Cindy Bell and Sergi Beltran and Christopher Benz and Andrew Berchuck and Bergmann, {Anke K.} and Bergstrom, {Erik N.} and Berman, {Benjamin P.} and Berney, {Daniel M.} and Bernhart, {Stephan H.} and Rameen Beroukhim and Mario Berrios and Samantha Bersani and Johanna Bertl and Miguel Betancourt and Vinayak Bhandari and Bhosle, {Shriram G.} and Biankin, {Andrew V.} and Matthias Bieg and Darell Bigner and Hans Binder and Ewan Birney and Michael Birrer and Biswas, {Nidhan K.} and Bodil Bjerkehagen and Tom Bodenheimer and Lori Boice and Giada Bonizzato and {De Bono}, {Johann S.} and Arnoud Boot and Bootwalla, {Moiz S.} and Ake Borg and Arndt Borkhardt and Boroevich, {Keith A.} and Ivan Borozan and Christoph Borst and Marcus Bosenberg and Mattia Bosio and Jacqueline Boultwood and Guillaume Bourque and Boutros, {Paul C.} and Bova, {G. Steven} and Bowen, {David T.} and Reanne Bowlby and Bowtell, {David D.L.} and Sandrine Boyault and Rich Boyce and Jeffrey Boyd and Alvis Brazma and Paul Brennan and Brewer, {Daniel S.} and Brinkman, {Arie B.} and Bristow, {Robert G.} and Broaddus, {Russell R.} and Brock, {Jane E.} and Malcolm Brock and Annegien Broeks and Brooks, {Angela N.} and Denise Brooks and Benedikt Brors and Li Ding and Fulton, {Lucinda A.} and Fulton, {Robert S.} and Ramaswamy Govindan and Reyka Jayasinghe and Tim Ley and Miller, {Christopher A.} and David Mutch and Wendl, {Michael C.}",

note = "Publisher Copyright: {\textcopyright} 2020, The Author(s).",

year = "2020",

month = dec,

day = "1",

doi = "10.1038/s41467-019-13825-8",

language = "English",

volume = "11",

journal = "Nature communications",

issn = "2041-1723",

number = "1",

}

PCAWG Tumor Subtypes and Clinical Translation Working Group, PCAWG Consortium, Jiao, W, Atwal, G, Polak, P, Karlic, R, Cuppen, E, Al-Shahrour, F, Atwal, G, Bailey, PJ, Biankin, AV, Boutros, PC, Campbell, PJ, Chang, DK, Cooke, SL, Deshpande, V, Faltas, BM, Faquin, WC, Garraway, L, Getz, G, Grimmond, SM, Haider, S, Hoadley, KA, Jiao, W, Kaiser, VB, Karlic, R, Kato, M, Kübler, K, Lazar, AJ, Li, CH, Louis, DN, Margolin, A, Martin, S, Nahal-Bose, HK, Nielsen, GP, Nik-Zainal, S, Omberg, L, P’ng, C, Perry, MD, Polak, P, Rheinbay, E, Rubin, MA, Semple, CA, Sgroi, DC, Shibata, T, Siebert, R, Smith, J, Stein, LD, Stobbe, MD, Sun, RX, Thai, K, Wright, DW, Wu, CL, Yuan, K, Zhang, J, Danyi, A, de Ridder, J, van Herpen, C, Lolkema, MP, Steeghs, N, Getz, G, Morris, QD, Aaltonen, LA, Abascal, F, Abeshouse, A, Aburatani, H, Adams, DJ, Agrawal, N, Ahn, KS, Ahn, SM, Aikata, H, Akbani, R, Akdemir, KC, Al-Ahmadie, H, Al-Sedairy, ST, Al-Shahrour, F, Alawi, M, Albert, M, Aldape, K, Alexandrov, LB, Ally, A, Alsop, K, Alvarez, EG, Amary, F, Amin, SB, Aminou, B, Ammerpohl, O, Anderson, MJ, Ang, Y, Antonello, D, Anur, P, Aparicio, S, Appelbaum, EL, Arai, Y, Aretz, A, Arihiro, K, Ariizumi, SI, Armenia, J, Arnould, L, Asa, S, Assenov, Y, Atwal, G, Aukema, S, Auman, JT, Aure, MRR, Awadalla, P, Aymerich, M, Bader, GD, Baez-Ortega, A, Bailey, MH, Bailey, PJ, Balasundaram, M, Balu, S, Bandopadhayay, P, Banks, RE, Barbi, S, Barbour, AP, Barenboim, J, Barnholtz-Sloan, J, Barr, H, Barrera, E, Bartlett, J, Bartolome, J, Bassi, C, Bathe, OF, Baumhoer, D, Bavi, P, Baylin, SB, Bazant, W, Beardsmore, D, Beck, TA, Behjati, S, Behren, A, Niu, B, Bell, C, Beltran, S, Benz, C, Berchuck, A, Bergmann, AK, Bergstrom, EN, Berman, BP, Berney, DM, Bernhart, SH, Beroukhim, R, Berrios, M, Bersani, S, Bertl, J, Betancourt, M, Bhandari, V, Bhosle, SG, Biankin, AV, Bieg, M, Bigner, D, Binder, H, Birney, E, Birrer, M, Biswas, NK, Bjerkehagen, B, Bodenheimer, T, Boice, L, Bonizzato, G, De Bono, JS, Boot, A, Bootwalla, MS, Borg, A, Borkhardt, A, Boroevich, KA, Borozan, I, Borst, C, Bosenberg, M, Bosio, M, Boultwood, J, Bourque, G, Boutros, PC, Bova, GS, Bowen, DT, Bowlby, R, Bowtell, DDL, Boyault, S, Boyce, R, Boyd, J, Brazma, A, Brennan, P, Brewer, DS, Brinkman, AB, Bristow, RG, Broaddus, RR, Brock, JE, Brock, M, Broeks, A, Brooks, AN, Brooks, D, Brors, B, Ding, L , Fulton, LA , Fulton, RS , Govindan, R , Jayasinghe, R , Ley, T , Miller, CA , Mutch, D & Wendl, MC 2020, 'A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns', Nature communications, vol. 11, no. 1, 728. https://doi.org/10.1038/s41467-019-13825-8

TY - JOUR

T1 - A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns

AU - PCAWG Tumor Subtypes and Clinical Translation Working Group

AU - PCAWG Consortium

AU - Jiao, Wei

AU - Atwal, Gurnit

AU - Polak, Paz

AU - Karlic, Rosa

AU - Cuppen, Edwin

AU - Al-Shahrour, Fatima

AU - Atwal, Gurnit

AU - Bailey, Peter J.

AU - Biankin, Andrew V.

AU - Boutros, Paul C.

AU - Campbell, Peter J.

AU - Chang, David K.

AU - Cooke, Susanna L.

AU - Deshpande, Vikram

AU - Faltas, Bishoy M.

AU - Faquin, William C.

AU - Garraway, Levi

AU - Getz, Gad

AU - Grimmond, Sean M.

AU - Haider, Syed

AU - Hoadley, Katherine A.

AU - Jiao, Wei

AU - Kaiser, Vera B.

AU - Karlic, Rosa

AU - Kato, Mamoru

AU - Kübler, Kirsten

AU - Lazar, Alexander J.

AU - Li, Constance H.

AU - Louis, David N.

AU - Margolin, Adam

AU - Martin, Sancha

AU - Nahal-Bose, Hardeep K.

AU - Nielsen, G. Petur

AU - Nik-Zainal, Serena

AU - Omberg, Larsson

AU - P’ng, Christine

AU - Perry, Marc D.

AU - Polak, Paz

AU - Rheinbay, Esther

AU - Rubin, Mark A.

AU - Semple, Colin A.

AU - Sgroi, Dennis C.

AU - Shibata, Tatsuhiro

AU - Siebert, Reiner

AU - Smith, Jaclyn

AU - Stein, Lincoln D.

AU - Stobbe, Miranda D.

AU - Sun, Ren X.

AU - Thai, Kevin

AU - Wright, Derek W.

AU - Wu, Chin Lee

AU - Yuan, Ke

AU - Zhang, Junjun

AU - Danyi, Alexandra

AU - de Ridder, Jeroen

AU - van Herpen, Carla

AU - Lolkema, Martijn P.

AU - Steeghs, Neeltje

AU - Getz, Gad

AU - Morris, Quaid D.

AU - Aaltonen, Lauri A.

AU - Abascal, Federico

AU - Abeshouse, Adam

AU - Aburatani, Hiroyuki

AU - Adams, David J.

AU - Agrawal, Nishant

AU - Ahn, Keun Soo

AU - Ahn, Sung Min

AU - Aikata, Hiroshi

AU - Akbani, Rehan

AU - Akdemir, Kadir C.

AU - Al-Ahmadie, Hikmat

AU - Al-Sedairy, Sultan T.

AU - Al-Shahrour, Fatima

AU - Alawi, Malik

AU - Albert, Monique

AU - Aldape, Kenneth

AU - Alexandrov, Ludmil B.

AU - Ally, Adrian

AU - Alsop, Kathryn

AU - Alvarez, Eva G.

AU - Amary, Fernanda

AU - Amin, Samirkumar B.

AU - Aminou, Brice

AU - Ammerpohl, Ole

AU - Anderson, Matthew J.

AU - Ang, Yeng

AU - Antonello, Davide

AU - Anur, Pavana

AU - Aparicio, Samuel

AU - Appelbaum, Elizabeth L.

AU - Arai, Yasuhito

AU - Aretz, Axel

AU - Arihiro, Koji

AU - Ariizumi, Shun ichi

AU - Armenia, Joshua

AU - Arnould, Laurent

AU - Asa, Sylvia

AU - Assenov, Yassen

AU - Atwal, Gurnit

AU - Aukema, Sietse

AU - Auman, J. Todd

AU - Aure, Miriam R.R.

AU - Awadalla, Philip

AU - Aymerich, Marta

AU - Bader, Gary D.

AU - Baez-Ortega, Adrian

AU - Bailey, Matthew H.

AU - Bailey, Peter J.

AU - Balasundaram, Miruna

AU - Balu, Saianand

AU - Bandopadhayay, Pratiti

AU - Banks, Rosamonde E.

AU - Barbi, Stefano

AU - Barbour, Andrew P.

AU - Barenboim, Jonathan

AU - Barnholtz-Sloan, Jill

AU - Barr, Hugh

AU - Barrera, Elisabet

AU - Bartlett, John

AU - Bartolome, Javier

AU - Bassi, Claudio

AU - Bathe, Oliver F.

AU - Baumhoer, Daniel

AU - Bavi, Prashant

AU - Baylin, Stephen B.

AU - Bazant, Wojciech

AU - Beardsmore, Duncan

AU - Beck, Timothy A.

AU - Behjati, Sam

AU - Behren, Andreas

AU - Niu, Beifang

AU - Bell, Cindy

AU - Beltran, Sergi

AU - Benz, Christopher

AU - Berchuck, Andrew

AU - Bergmann, Anke K.

AU - Bergstrom, Erik N.

AU - Berman, Benjamin P.

AU - Berney, Daniel M.

AU - Bernhart, Stephan H.

AU - Beroukhim, Rameen

AU - Berrios, Mario

AU - Bersani, Samantha

AU - Bertl, Johanna

AU - Betancourt, Miguel

AU - Bhandari, Vinayak

AU - Bhosle, Shriram G.

AU - Biankin, Andrew V.

AU - Bieg, Matthias

AU - Bigner, Darell

AU - Binder, Hans

AU - Birney, Ewan

AU - Birrer, Michael

AU - Biswas, Nidhan K.

AU - Bjerkehagen, Bodil

AU - Bodenheimer, Tom

AU - Boice, Lori

AU - Bonizzato, Giada

AU - De Bono, Johann S.

AU - Boot, Arnoud

AU - Bootwalla, Moiz S.

AU - Borg, Ake

AU - Borkhardt, Arndt

AU - Boroevich, Keith A.

AU - Borozan, Ivan

AU - Borst, Christoph

AU - Bosenberg, Marcus

AU - Bosio, Mattia

AU - Boultwood, Jacqueline

AU - Bourque, Guillaume

AU - Boutros, Paul C.

AU - Bova, G. Steven

AU - Bowen, David T.

AU - Bowlby, Reanne

AU - Bowtell, David D.L.

AU - Boyault, Sandrine

AU - Boyce, Rich

AU - Boyd, Jeffrey

AU - Brazma, Alvis

AU - Brennan, Paul

AU - Brewer, Daniel S.

AU - Brinkman, Arie B.

AU - Bristow, Robert G.

AU - Broaddus, Russell R.

AU - Brock, Jane E.

AU - Brock, Malcolm

AU - Broeks, Annegien

AU - Brooks, Angela N.

AU - Brooks, Denise

AU - Brors, Benedikt

AU - Ding, Li

AU - Fulton, Lucinda A.

AU - Fulton, Robert S.

AU - Govindan, Ramaswamy

AU - Jayasinghe, Reyka

AU - Ley, Tim

AU - Miller, Christopher A.

AU - Mutch, David

AU - Wendl, Michael C.

PY - 2020/12/1

Y1 - 2020/12/1

N2 - In cancer, the primary tumour’s organ of origin and histopathology are the strongest determinants of its clinical behaviour, but in 3% of cases a patient presents with a metastatic tumour and no obvious primary. Here,as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, we train a deep learning classifier to predict cancer type based on patterns of somatic passenger mutations detected in whole genome sequencing (WGS) of 2606 tumours representing 24 common cancer types produced by the PCAWG Consortium. Our classifier achieves an accuracy of 91% on held-out tumor samples and 88% and 83% respectively on independent primary and metastatic samples, roughly double the accuracy of trained pathologists when presented with a metastatic tumour without knowledge of the primary. Surprisingly, adding information on driver mutations reduced accuracy. Our results have clinical applicability, underscore how patterns of somatic passenger mutations encode the state of the cell of origin, and can inform future strategies to detect the source of circulating tumour DNA.

AB - In cancer, the primary tumour’s organ of origin and histopathology are the strongest determinants of its clinical behaviour, but in 3% of cases a patient presents with a metastatic tumour and no obvious primary. Here,as part of the ICGC/TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, we train a deep learning classifier to predict cancer type based on patterns of somatic passenger mutations detected in whole genome sequencing (WGS) of 2606 tumours representing 24 common cancer types produced by the PCAWG Consortium. Our classifier achieves an accuracy of 91% on held-out tumor samples and 88% and 83% respectively on independent primary and metastatic samples, roughly double the accuracy of trained pathologists when presented with a metastatic tumour without knowledge of the primary. Surprisingly, adding information on driver mutations reduced accuracy. Our results have clinical applicability, underscore how patterns of somatic passenger mutations encode the state of the cell of origin, and can inform future strategies to detect the source of circulating tumour DNA.

UR - http://www.scopus.com/inward/record.url?scp=85079062177&partnerID=8YFLogxK

U2 - 10.1038/s41467-019-13825-8

DO - 10.1038/s41467-019-13825-8

M3 - Article

C2 - 32024849

AN - SCOPUS:85079062177

SN - 2041-1723

VL - 11

JO - Nature communications

JF - Nature communications

IS - 1

M1 - 728

ER -

A deep learning system accurately classifies primary and metastatic cancers using passenger mutation patterns

Abstract

Access to Document

Other files and links

Fingerprint

Cite this