biostudies-literatureMourikis TPCancer Research UKMedical Research CouncilThe Francis Crick InstituteEngineering and Physical Sciences Research Council3101https://www.ebi.ac.uk/biostudies/studies/S-EPMC6629660biostudies-literatureUnknown10(1)The identification of cancer-promoting genetic alterations is challenging particularly in highly unstable and heterogeneous cancers, such as esophageal adenocarcinoma (EAC). Here we describe a machine learning algorithm to identify cancer genes in individual patients considering all types of damaging alterations simultaneously. Analysing 261 EACs from the OCCAMS Consortium, we discover helper genes that, alongside well-known drivers, promote cancer. We confirm the robustness of our approach in 107 additional EACs. Unlike recurrent alterations of known drivers, these cancer helper genes are rare or patient-specific. However, they converge towards perturbations of well-known cancer processes. Recurrence of the same process perturbations, rather than individual genes, divides EACs into six clNature communicationsPatient-specific cancer genes contribute to recurrently perturbed pathways and establish therapeutic vulnerabilities in esophageal adenocarcinoma.PMC6629660100082392323924MC_PC_160481786366MR/L001411/110152S_3659164632548720406C43634/A25487Davies ALynch AGde la Rue RCereda MContino GLovat LYau CRobert O'Neill JHughes CTaniere PLishman SMahadeva USujendran VCheong ESanders GGrehan NWalker RCDavies JKaye PO'Donovan MFitzgerald RCBornschein JOakes SIgali LPeters CJJammula SFickling WHayes SJMoorthy KMacRae SSecrier MElliott RFCrichton CPuig SHupp TRShepherd NBenedetti LParsons SLChang FKatz-Summercorn ACrawte JMiremadi ATemelkovski DBarr HPreston SRCiccarelli FDEdwards PAWTurkington RFoxall ESafranek PCarroll NLagergren JGoh VSaunders JHanna GBHowell MTavare SNoorani ASave VLewis MOesophageal Cancer Clinical and Molecular Stratification (OCCAMS) ConsortiumDevonshire GNutzinger BLi XMcManus DAng YLoureda DGrabowska AMalhotra SMourikis TPBower LHindmarsh AScott MFidziukiewicz ESothi SGossage JPerner JScaffidi PZylstra JHaidry RSuortamo SHarden CSkipworth RJEBeggs AUnderwood TJBagwan IAbbas SNulsen JBerrisford RTripathi MKhoo DSharrocks AGrace BLSoomro INorthrop ATucker OOld OKumar BEldridge MfalsePatient-specific cancer genes contribute to recurrently perturbed pathways and establish therapeutic vulnerabilities in esophageal adenocarcinoma.The identification of cancer-promoting genetic alterations is challenging particularly in highly unstable and heterogeneous cancers, such as esophageal adenocarcinoma (EAC). Here we describe a machine learning algorithm to identify cancer genes in individual patients considering all types of damaging alterations simultaneously. Analysing 261 EACs from the OCCAMS Consortium, we discover helper genes that, alongside well-known drivers, promote cancer. We confirm the robustness of our approach in 107 additional EACs. Unlike recurrent alterations of known drivers, these cancer helper genes are rare or patient-specific. However, they converge towards perturbations of well-known cancer processes. Recurrence of the same process perturbations, rather than individual genes, divides EACs into six cl2019-01-01T00:00:00Z2019 Jul2026-05-06T21:56:17.78Z2019-07-25T07:10:08ZS-EPMC66296603130837710.1038/s41467-019-10898-3