In this section
@article{Ashdown:2020:10.1126/sciadv.aba9338,
author = {Ashdown, G and Gaboriau, D and Baum, J},
doi = {10.1126/sciadv.aba9338},
journal = {Science Advances},
title = {A machine learning approach to define antimalarial drug action from heterogeneous cell-based screens},
url = {http://dx.doi.org/10.1126/sciadv.aba9338},
volume = {6},
year = {2020}
}
TY - JOUR
AB - Drug resistance threatens the effective prevention and treatment of an ever-increasing range ofhuman infections. This highlights an urgent need for new and improved drugs with novelmechanisms of action to avoid cross-resistance. Current cell-based drug screens are,however, restricted to binary live/dead readouts with no provision for mechanism of actionprediction. Machine learning methods are increasingly being used to improve informationextraction from imaging data. Such methods, however, work poorly with heterogeneouscellular phenotypes and generally require time-consuming human-led training. We havedeveloped a semi-supervised machine learning approach, combining human- and machinelabelled training data from mixed human malaria parasite cultures. Designed for highthroughput and high-resolution screening, our semi-supervised approach is robust to naturalparasite morphological heterogeneity and correctly orders parasite developmental stages. Ourapproach also reproducibly detects and clusters drug-induced morphological outliers bymechanism of action, demonstrating the potential power of machine learning for acceleratingcell-based drug discovery.
AU - Ashdown,G
AU - Gaboriau,D
AU - Baum,J
DO - 10.1126/sciadv.aba9338
PY - 2020///
SN - 2375-2548
TI - A machine learning approach to define antimalarial drug action from heterogeneous cell-based screens
T2 - Science Advances
UR - http://dx.doi.org/10.1126/sciadv.aba9338
UR - http://hdl.handle.net/10044/1/80889
VL - 6
ER -
Join our mailing list to receive updates about network news and events and to connect with other members
(Imperial staff and students only)
