Two-step multi-omics modelling of drug sensitivity in cancer cell lines to identify driving mechanisms

被引:6
|
作者
Kusch, Nina [1 ,2 ,3 ]
Schuppert, Andreas [1 ,2 ,3 ]
机构
[1] Rhein Westfal TH Aachen, Joint Res Ctr Computat Biomed, Aachen, Germany
[2] Rhein Westfal TH Aachen, Aachen Inst Adv Study Computat Engn Sci AICES, Aachen, Germany
[3] Uniklin Aachen, Aachen, Germany
来源
PLOS ONE | 2020年 / 15卷 / 11期
关键词
P53; MDM2; OVEREXPRESSION; DOCETAXEL; LYMPHOMA;
D O I
10.1371/journal.pone.0238961
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
Drug sensitivity prediction models for human cancer cell lines constitute important tools in identifying potential computational biomarkers for responsiveness in a pre-clinical setting. Integrating information derived from a range of heterogeneous data is crucial, but remains non-trivial, as differences in data structures may hinder fitting algorithms from assigning adequate weights to complementary information that is contained in distinct omics data. In order to counteract this effect that tends to lead to just one data type dominating supposedly multi-omics models, we developed a novel tool that enables users to train single-omics models separately in a first step and to integrate them into a multi-omics model in a second step. Extensive ablation studies are performed in order to facilitate an in-depth evaluation of the respective contributions of singular data types and of combinations thereof, effectively identifying redundancies and interdependencies between them. Moreover, the integration of the single-omics models is realized by a range of distinct classification algorithms, thus allowing for a performance comparison. Sets of molecular events and tissue types found to be related to significant shifts in drug sensitivity are returned to facilitate a comprehensive and straightforward analysis of potential computational biomarkers for drug responsiveness. Our two-step approach yields sets of actual multi-omics pan-cancer classification models that are highly predictive for a majority of drugs in the GDSC data base. In the context of targeted drugs with particular modes of action, its predictive performances compare favourably to those of classification models that incorporate multi-omics data in a simple one-step approach. Additionally, case studies demonstrate that it succeeds both in correctly identifying known key biomarkers for sensitivity towards specific drug compounds as well as in providing sets of potential candidates for additional computational biomarkers.
引用
收藏
页数:22
相关论文
共 50 条
  • [1] Multi-omics analysis to identify driving factors in colorectal cancer
    Xu, Xi
    Gong, Chaoju
    Wang, Yunfeng
    Hu, Yanyan
    Liu, Hong
    Fang, Zejun
    EPIGENOMICS, 2020, 12 (18) : 1633 - 1650
  • [2] Multi-omics data of gastric cancer cell lines
    Seo, Eun-Hye
    Shin, Yun-Jae
    Kim, Hee-Jin
    Kim, Jeong-Hwan
    Kim, Yong Sung
    Kim, Seon-Young
    BMC GENOMIC DATA, 2023, 24 (01):
  • [3] Multi-omics data of gastric cancer cell lines
    Eun-Hye Seo
    Yun-Jae Shin
    Hee-Jin Kim
    Jeong-Hwan Kim
    Yong Sung Kim
    Seon-Young Kim
    BMC Genomic Data, 24
  • [4] An integrated multi-omics approach to identify regulatory mechanisms in cancer metastatic processes
    Saba Ghaffari
    Casey Hanson
    Remington E. Schmidt
    Kelly J. Bouchonville
    Steven M. Offer
    Saurabh Sinha
    Genome Biology, 22
  • [5] An integrated multi-omics approach to identify regulatory mechanisms in cancer metastatic processes
    Ghaffari, Saba
    Hanson, Casey
    Schmidt, Remington E.
    Bouchonville, Kelly J.
    Offer, Steven M.
    Sinha, Saurabh
    GENOME BIOLOGY, 2021, 22 (01)
  • [6] Statistical integration of multi-omics and drug screening data from cell lines
    el Bouhaddani, Said
    Hoellerhage, Matthias
    Uh, Hae-Won
    Moebius, Claudia
    Bickle, Marc
    Hoeglinger, Gunter
    Houwing-Duistermaat, Jeanine
    PLOS COMPUTATIONAL BIOLOGY, 2024, 20 (01)
  • [7] Multi-omics of 34 colorectal cancer cell lines - a resource for biomedical studies
    Kaja C. G. Berg
    Peter W. Eide
    Ina A. Eilertsen
    Bjarne Johannessen
    Jarle Bruun
    Stine A. Danielsen
    Merete Bjørnslett
    Leonardo A. Meza-Zepeda
    Mette Eknæs
    Guro E. Lind
    Ola Myklebost
    Rolf I. Skotheim
    Anita Sveen
    Ragnhild A. Lothe
    Molecular Cancer, 16
  • [8] Multi-omics of 34 colorectal cancer cell lines - a resource for biomedical studies
    Berg, Kaja C. G.
    Eide, Peter W.
    Eilertsen, Ina A.
    Johannessen, Bjarne
    Bruun, Jarle
    Danielsen, Stine A.
    Bjornslett, Merete
    Meza-Zepeda, Leonardo A.
    Eknaes, Mette
    Lind, Guro E.
    Myklebost, Ola
    Skotheim, Rolf I.
    Sveen, Anita
    Lothe, Ragnhild A.
    MOLECULAR CANCER, 2017, 16
  • [9] Multi-omics analysis of mechanisms underlying sensitivity of Merkel cell carcinoma to pyrvinium pamoate
    Yang, Jiawen
    Lim, James T.
    Victor, Paul
    Schnellmann, Rick G.
    DeCaprio, James A.
    Padi, Megha
    CANCER RESEARCH, 2023, 83 (07)
  • [10] Divergent organ-specific isogenic metastatic cell lines identified using multi-omics exhibit differential drug sensitivity
    Winnard, Paul T., Jr.
    Vesuna, Farhad
    Muthukumar, Sankar
    Raman, Venu
    PLOS ONE, 2020, 15 (11):