Sharif Digital Repository

Single-cell resolution technologies warrant computational methods that capture cell heterogeneity while allowing efficient comparisons of populations. Here, we summarize cell populations by adding features’ dispersion and covariances to population averages, in the context of image-based profiling. We find that data fusion is critical for these metrics to improve results over the prior alternatives, providing at least ~20% better performance in predicting a compound’s mechanism of action (MoA) and a gene’s pathway. © 2019, The Author(s)  

The pharmaceutical industry is facing enormous challenges due to high drug attribution rates. For the past decades, novel methods have been developed for safety and efficacy testing, as well as for improving early development stages. In vitro screening methods for drug-receptor binding are considered to be good alternatives for decreasing costs in the identification of drug candidates. However, these methods require lengthy and troublesome labeling steps. Biosensors hold great promise due to the fact that label-free detection schemes can be designed in an easy and low-cost manner. In this paper, for the first time in the literature, we aimed to compare the potential of label-free optical and...