TY - JOUR
T1 - Few-shot learning creates predictive models of drug response that translate from high-throughput screens to individual patients
AU - Ma , Jianzhu
AU - Fong, Samson H.
AU - Luo, Yunan
AU - Bakkenist, Christopher J.
AU - Shen, John Paul
AU - Mourragui, Soufiane
AU - Wessels, Lodewyk F.A.
AU - Hafner, Marc
AU - Sharan, Roded
AU - Jiang, Peng
AU - Ideker, Trey
PY - 2021
Y1 - 2021
N2 - Cell-line screens create expansive datasets for learning predictive markers of drug response, but these models do not readily translate to the clinic with its diverse contexts and limited data. In the present study, we apply a recently developed technique, few-shot machine learning, to train a versatile neural network model in cell lines that can be tuned to new contexts using few additional samples. The model quickly adapts when switching among different tissue types and in moving from cell-line models to clinical contexts, including patient-derived tumor cells and patient-derived xenografts. It can also be interpreted to identify the molecular features most important to a drug response, highlighting critical roles for RB1 and SMAD4 in the response to CDK inhibition and RNF8 and CHD4 in the response to ATM inhibition. The few-shot learning framework provides a bridge from the many samples surveyed in high-throughput screens (n-of-many) to the distinctive contexts of individual patients (n-of-one).
AB - Cell-line screens create expansive datasets for learning predictive markers of drug response, but these models do not readily translate to the clinic with its diverse contexts and limited data. In the present study, we apply a recently developed technique, few-shot machine learning, to train a versatile neural network model in cell lines that can be tuned to new contexts using few additional samples. The model quickly adapts when switching among different tissue types and in moving from cell-line models to clinical contexts, including patient-derived tumor cells and patient-derived xenografts. It can also be interpreted to identify the molecular features most important to a drug response, highlighting critical roles for RB1 and SMAD4 in the response to CDK inhibition and RNF8 and CHD4 in the response to ATM inhibition. The few-shot learning framework provides a bridge from the many samples surveyed in high-throughput screens (n-of-many) to the distinctive contexts of individual patients (n-of-one).
UR - http://www.scopus.com/inward/record.url?scp=85099938332&partnerID=8YFLogxK
U2 - 10.1038/s43018-020-00169-2
DO - 10.1038/s43018-020-00169-2
M3 - Article
AN - SCOPUS:85099938332
SN - 2662-1347
VL - 2
SP - 233
EP - 244
JO - Nature Cancer
JF - Nature Cancer
IS - 2
ER -