Machine Learning Based Prediction of Enzymatic Degradation of Plastics Using Encoded Protein Sequence and Effective Feature Representation

Abstract

Enzyme biocatalysis for plastic treatment and recycling is an emerging field of growing interest. However, it is challenging and time-consuming to identify plastic-degrading enzymes with desirable functionality, given the large number of putative enzyme sequences. There is a critical need to develop an effective approach to accurately predict the enzyme activity in degrading different types of plastics. In this study, we developed a machine-learning-based plastic enzymatic degradation (PED) framework to predict the ability of an enzyme to degrade plastics of interest by exploring and recognizing hidden patterns in protein sequences. A data set integrating information from a wide range of experimentally verified enzymes and various common plastic substrates was created. A new context-aware enzyme sequence representation (CESR) mechanism was developed to learn the abundant contextual information in enzyme sequences, and feature extraction was performed for enzymes at both the amino acid level and global sequence level. Thirteen machine learning classification algorithms were compared, and XGBoost was identified as the best-performing algorithm. PED achieved an overall accuracy of 90.2% and outperformed sequence-based protein classification models from the existing literature. Furthermore, important enzyme features in plastic degradation were identified and comprehensively interpreted. This study demonstrated a new tool for the prediction and discovery of plastic-degrading enzymes.

Toward a Circular Bioeconomy: Designing Microbes and Polymers for Biodegradation

“…Recently, machine learning was implemented to develop the plastic enzymatic degradation (PED) framework . This PED framework can predict the ability of an enzyme to degrade plastics by analyzing protein sequences to identify specific patterns.…”

“…Recently, machine learning was implemented to develop the plastic enzymatic degradation (PED) framework. 117 This PED framework can predict the ability of an enzyme to degrade plastics by analyzing protein sequences to identify specific patterns. The framework outperformed many other sequence-based classification models and identified that hydrophobicity, heat capacity, frequency of occurrence of alanine in the sequence, and the enzyme’s molecular weight are the most important structural motifs for an enzyme to degrade plastics.…”

“…The framework outperformed many other sequence-based classification models and identified that hydrophobicity, heat capacity, frequency of occurrence of alanine in the sequence, and the enzyme’s molecular weight are the most important structural motifs for an enzyme to degrade plastics. 117 This PED framework works by inputting an enzyme sequence and plastic type, and the framework outputs “degradable” or “nondegradable” with an accuracy of 90.2% (tested with 230 enzyme–plastic pairs from 129 enzyme sequences and 11 types of plastics). 117 …”

Enhancing PET Degrading Enzymes: A Combinatory Approach

“…[13] Similarly, MutCompute was applied to TfCut2 wild type cutinase to identify beneficial mutations and create an enhanced variant, which exhibited a 5.3-fold improved depolymerization of crystalline PET. [31]] Most recent examples include the adoption of machine learning-assisted prediction of the degrading activity of certain plastics [32] or the use of machine learning to guide the directed evolution of plastic degrading enzymes. [33] This paper has applied different approaches such as structure-based design, ancestral sequence reconstruction and machine learning to engineer a new variant of PsPETase.…”

Enzymatic degradation of polylactic acid (PLA)

“…In addition to the improvement of known enzymes, new PLA-degrading enzymes, with desirable properties, can be discovered using novel bioinformatic computational prediction methods. As sequence similarity does not consistently correlate with the functional ability of enzymes to degrade plastics, a machine learning-based approach was recently developed for the prediction of such enzymes (Jiang et al 2023 ). The system integrated data of a wide range of plastic-degrading enzymes, including PLA depolymerases like CLE and PlaM4, and could successfully establish sequence-based protein classification models.…”