Predicting the Glass Transition Temperature of Biopolymers via High-Throughput Molecular Dynamics Simulations and Machine Learning

Nature has only provided us with a limited number of
biobased and
biodegradable building blocks. Therefore, the fine-tuning of the sustainable
polymer properties is expected to be achieved through the control
of the composition of biobased copolymers for targeted applications
such as cosmetics. Until now, the main approaches to alleviate the
experimental efforts and accelerate the discovery of polymers have
relied on machine learning models trained on experimental data, which
implies enormous and difficult work in the compilation of data from
heterogeneous sources. On the other hand, molecular dynamics simulations
of polymers have shown that they can accurately capture the experimental
trends for a series of properties. However, the combination of different
ratios of monomers in copolymers can rapidly lead to a combinatorial
explosion, preventing investigation of all possibilities via molecular
dynamics simulations. In this work, we show that the combination of
machine learning approaches and high-throughput molecular dynamics
simulations permits quick and efficient sampling and characterization
of the relevant chemical design space for specific applications. Reliable
simulation protocols have been implemented to evaluate the glass transition
temperature of a series of 58 homopolymers, which exhibit good agreement
with experiments, and 488 copolymers. Overall, 2,184 simulations (four
replicas per polymer) were performed, for a total simulation time
of 143.052 μs. These results, constituting a data set of 546
polymers, have been used to train a machine learning model for the
prediction of the MD-calculated glass transition temperature with
a mean absolute error of 19.34 K and an R2 score of 0.83. Overall, within its applicability domain,
this machine learning model provides an impressive acceleration over
molecular dynamics simulations: the glass transition temperature of
thousands of polymers can be obtained within seconds, whereas it would
have taken node-years to simulate them. This type of approach can
be tuned to address different design spaces or different polymer properties
and thus has the potential to accelerate the discovery of polymers.