Your browser doesn't support javascript.
loading
Privacy preserving validation for multiomic prediction models.
Ahmed, Talal; Carty, Mark A; Wenric, Stephane; Dry, Jonathan R; Salahudeen, Ameen A; Khan, Aly A; Lefkofsky, Eric; Stumpe, Martin C; Pelossof, Raphael.
Afiliação
  • Ahmed T; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Carty MA; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Wenric S; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Dry JR; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Salahudeen AA; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Khan AA; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Lefkofsky E; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Stumpe MC; Tempus Labs Inc., Chicago, IL 60654, USA.
  • Pelossof R; Tempus Labs Inc., Chicago, IL 60654, USA.
Brief Bioinform ; 23(3)2022 05 13.
Article em En | MEDLINE | ID: mdl-35388408
Reproducibility of results obtained using ribonucleic acid (RNA) data across labs remains a major hurdle in cancer research. Often, molecular predictors trained on one dataset cannot be applied to another due to differences in RNA library preparation and quantification, which inhibits the validation of predictors across labs. While current RNA correction algorithms reduce these differences, they require simultaneous access to patient-level data from all datasets, which necessitates the sharing of training data for predictors when sharing predictors. Here, we describe SpinAdapt, an unsupervised RNA correction algorithm that enables the transfer of molecular models without requiring access to patient-level data. It computes data corrections only via aggregate statistics of each dataset, thereby maintaining patient data privacy. Despite an inherent trade-off between privacy and performance, SpinAdapt outperforms current correction methods, like Seurat and ComBat, on publicly available cancer studies, including TCGA and ICGC. Furthermore, SpinAdapt can correct new samples, thereby enabling unbiased evaluation on validation cohorts. We expect this novel correction paradigm to enhance research reproducibility and to preserve patient privacy.
Assuntos
Palavras-chave

Texto completo: 1 Bases de dados: MEDLINE Assunto principal: Confidencialidade / Privacidade Tipo de estudo: Prognostic_studies / Risk_factors_studies Limite: Humans Idioma: En Revista: Brief Bioinform Assunto da revista: BIOLOGIA / INFORMATICA MEDICA Ano de publicação: 2022 Tipo de documento: Article País de afiliação: Estados Unidos

Texto completo: 1 Bases de dados: MEDLINE Assunto principal: Confidencialidade / Privacidade Tipo de estudo: Prognostic_studies / Risk_factors_studies Limite: Humans Idioma: En Revista: Brief Bioinform Assunto da revista: BIOLOGIA / INFORMATICA MEDICA Ano de publicação: 2022 Tipo de documento: Article País de afiliação: Estados Unidos