Integrating multimodal data sets into a mathematical framework to describe and predict therapeutic resistance in cancer
| Autor: | Grant Howard, Angela M. Jarrett, Eric Brenner, Amy Brock, Aziz Al'Khafaji, Thomas E. Yankeelov, Daylin Morgan, William Mo, Kaitlyn E. Johnson, Russell E. Durrett, Andrea Gardner, Eduardo D. Sontag |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
0303 health sciences
Mathematical model business.industry Estimation theory Computer science Cancer Machine learning computer.software_genre medicine.disease Field (computer science) 3. Good health Data set Transcriptome 03 medical and health sciences Identification (information) 0302 clinical medicine Single-cell analysis 030220 oncology & carcinogenesis medicine Artificial intelligence business computer Biomedicine 030304 developmental biology |
| DOI: | 10.1101/2020.02.11.943738 |
| Popis: | SummaryA significant challenge in the field of biomedicine is the development of methods to integrate the multitude of dispersed data sets into comprehensive frameworks to be used to generate optimal clinical decisions. Recent technological advances in single cell analysis allow for high-dimensional molecular characterization of cells and populations, but to date, few mathematical models have attempted to integrate measurements from the single cell scale with other data types. Here, we present a framework that actionizes static outputs from a machine learning model and leverages these as measurements of state variables in a dynamic mechanistic model of treatment response. We apply this framework to breast cancer cells to integrate single cell transcriptomic data with longitudinal population-size data. We demonstrate that the explicit inclusion of the transcriptomic information in the parameter estimation is critical for identification of the model parameters and enables accurate prediction of new treatment regimens. Inclusion of the transcriptomic data improves predictive accuracy in new treatment response dynamics with a concordance correlation coefficient (CCC) of 0.89 compared to a prediction accuracy of CCC = 0.79 without integration of the single cell RNA sequencing (scRNA-seq) data directly into the model calibration. To the best our knowledge, this is the first work that explicitly integrates single cell clonally-resolved transcriptome datasets with longitudinal treatment response data into a mechanistic mathematical model of drug resistance dynamics. We anticipate this approach to be a first step that demonstrates the feasibility of incorporating multimodal data sets into identifiable mathematical models to develop optimized treatment regimens from data. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|