TY - JOUR
T1 - Predicting Proteolysis in Complex Proteomes Using Deep Learning
AU - Ozols, Matiss
AU - Eckersley, Alexander
AU - Platt, Christopher I.
AU - Stewart-McGuinness, Callum
AU - Hibbert, Sarah A.
AU - Revote, Jerico
AU - Li, Fuyi
AU - Griffiths, Christopher E. M.
AU - Watson, Rachel E. B.
AU - Song, Jiangning
AU - Bell, Mike
AU - Sherratt, Michael J.
N1 - Funding Information:
Funding: This research was funded by Walgreens Boots Alliance, Nottingham, UK. C.E.M.G. and R.E.B.W. are funded in part by the NIHR Manchester Biomedical Research Centre.
Publisher Copyright:
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2021/3/17
Y1 - 2021/3/17
N2 - Both protease-and reactive oxygen species (ROS)-mediated proteolysis are thought to be key effectors of tissue remodeling. We have previously shown that comparison of amino acid composition can predict the differential susceptibilities of proteins to photo-oxidation. However, predicting protein susceptibility to endogenous proteases remains challenging. Here, we aim to develop bioin-formatics tools to (i) predict cleavage site locations (and hence putative protein susceptibilities) and (ii) compare the predicted vulnerabilities of skin proteins to protease-and ROS-mediated proteolysis. The first goal of this study was to experimentally evaluate the ability of existing protease cleavage site prediction models (PROSPER and DeepCleave) to identify experimentally determined MMP9 cleavage sites in two purified proteins and in a complex human dermal fibroblast-derived extracellular matrix (ECM) proteome. We subsequently developed deep bidirectional recurrent neural network (BRNN) models to predict cleavage sites for 14 tissue proteases. The predictions of the new models were tested against experimental datasets and combined with amino acid composition analysis (to predict ultraviolet radiation (UVR)/ROS susceptibility) in a new web app: the Manchester proteome susceptibility calculator (MPSC). The BRNN models performed better in predicting cleavage sites in native dermal ECM proteins than existing models (DeepCleave and PROSPER), and application of MPSC to the skin proteome suggests that: compared with the elastic fiber network, fibrillar collagens may be susceptible primarily to protease-mediated proteolysis. We also identify additional putative targets of oxidative damage (dermatopontin, fibulins and defensins) and protease action (laminins and nidogen). MPSC has the potential to identify potential targets of proteolysis in disparate tissues and disease states.
AB - Both protease-and reactive oxygen species (ROS)-mediated proteolysis are thought to be key effectors of tissue remodeling. We have previously shown that comparison of amino acid composition can predict the differential susceptibilities of proteins to photo-oxidation. However, predicting protein susceptibility to endogenous proteases remains challenging. Here, we aim to develop bioin-formatics tools to (i) predict cleavage site locations (and hence putative protein susceptibilities) and (ii) compare the predicted vulnerabilities of skin proteins to protease-and ROS-mediated proteolysis. The first goal of this study was to experimentally evaluate the ability of existing protease cleavage site prediction models (PROSPER and DeepCleave) to identify experimentally determined MMP9 cleavage sites in two purified proteins and in a complex human dermal fibroblast-derived extracellular matrix (ECM) proteome. We subsequently developed deep bidirectional recurrent neural network (BRNN) models to predict cleavage sites for 14 tissue proteases. The predictions of the new models were tested against experimental datasets and combined with amino acid composition analysis (to predict ultraviolet radiation (UVR)/ROS susceptibility) in a new web app: the Manchester proteome susceptibility calculator (MPSC). The BRNN models performed better in predicting cleavage sites in native dermal ECM proteins than existing models (DeepCleave and PROSPER), and application of MPSC to the skin proteome suggests that: compared with the elastic fiber network, fibrillar collagens may be susceptible primarily to protease-mediated proteolysis. We also identify additional putative targets of oxidative damage (dermatopontin, fibulins and defensins) and protease action (laminins and nidogen). MPSC has the potential to identify potential targets of proteolysis in disparate tissues and disease states.
KW - Aging
KW - Biomark-ers
KW - Deep-learning
KW - Degradomics
KW - Extracellular matrix
KW - Machine learning
KW - Protease
KW - Skin
U2 - 10.3390/ijms22063071
DO - 10.3390/ijms22063071
M3 - Article
C2 - 33803033
VL - 22
SP - 1
EP - 20
JO - International Journal of Molecular Sciences
JF - International Journal of Molecular Sciences
SN - 1661-6596
IS - 6
M1 - 3071
ER -