Prediction and Design of Nanozymes using Explainable Machine Learning

Abstract

the intrinsic characteristics of nano materials, nanozymes have potential widespread applications within the fields of biosensing, [2] antibacterials, [3] environ mental pollution, [4] and disease therapy. [5] Since our discovery of ferromagnetic nanoparticles with intrinsic peroxidase like activity in 2007, [6] there have been thousands of publications that reported on enzymemimicking activities of nanoma terials, which involve at least six classes of enzymemimicry. [7] According to the litera ture, different nanomaterials can intrinsi cally possess the same enzymemimicking activities, [8] and certain types of nano materials tend to exhibit differential enzymelike catalytic activities. [9] The het erogeneous results reveal the complexity and diversity of nanozymes in terms of catalytic capacity. [10] Indeed, the particle property relationship of nanozymes is complicated, with a current lack of fun damental understanding. Furthermore, the synthesis of nanozymes with desired characteristics are generally determined by trial and error, and based on intui tion and experience, which are timeconsuming, laborious and resourceintensive.As a branch of artificial intelligence, machine learning aims to develop computational algorithms to infer mathematical An abundant number of nanomaterials have been discovered to possess enzyme-like catalytic activity, termed nanozymes. It is identified that a variety of internal and external factors influence the catalytic activity of nanozymes. However, there is a lack of essential methodologies to uncover the hidden mechanisms between nanozyme features and enzyme-like activity. Here, a data-driven approach is demonstrated that utilizes machine-learning algorithms to understand particle-property relationships, allowing for classification and quantitative predictions of enzyme-like activity exhibited by nanozymes. High consistency between predicted outputs and the observations is confirmed by accuracy (90.6%) and R 2 (up to 0.80). Furthermore, sensitive analysis of the models reveals the central roles of transition metals in determining nanozyme activity. As an example, the models are successfully applied to predict or design desirable nanozymes by uncovering the hidden relationship between different periods of transition metals and their enzyme-like performance. This study offers a promising strategy to develop nanozymes with desirable catalytic activity and demonstrates the potential of machine learning within the field of material science.The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/adma.202201736.

Machine Learning Assisted Graphdiyne-Based Nanozyme Discovery

“…27 Huang et al used deep neural network (DNN) algorithms to classify and quantitatively predict the enzyme-like activity exhibited by nanozymes and provided a promising strategy for developing nanozymes with desirable catalytic activity. 28 Despite their effectiveness in predicting nanozyme properties, the use of ML algorithms to guide nanozyme discovery is still fairly limited.…”

“…Razlivina et al established an open access database of nanozymes from over 100 published research papers and achieved high accuracy predictions of K m and K cat by the method of random forest . Huang et al used deep neural network (DNN) algorithms to classify and quantitatively predict the enzyme-like activity exhibited by nanozymes and provided a promising strategy for developing nanozymes with desirable catalytic activity . Despite their effectiveness in predicting nanozyme properties, the use of ML algorithms to guide nanozyme discovery is still fairly limited.…”

A User’s Guide to Machine Learning for Polymeric Biomaterials

“…Observation/dependence plots are also generated to interpret how different features interact with each other while influencing model performance. SHAP has been used to interrogate predictions of high performing polycationic delivery vehicles, to facilitate feature engineering for models of immunomodulatory osteoinductive biomaterials, and to conduct a sensitivity analysis for models predicting the enzymatic activity of synthetic nanomaterials . SHAP calculation can be implemented using the Python package shap, works well with tree-based models in scikit-learn, and is already implemented in several boosted-tree frameworks.…”

Machine Learning Guided Discovery of Superoxide Dismutase Nanozymes for Androgenetic Alopecia

“…Machine learning (ML) is a type of artificial intelligence that has gained extensive attention as a universal tool due to its exceptional performance in processing massive data efficiently in a variety of fields. − More importantly, ML has shown a strong ability to accelerate the discovery of new materials by effectively learning from a large amount of data, , but so far there are few studies regarding the ML assisted exploration of nanozymes. , Hence, it is highly desirable to accelerate the discovery of SOD mimics by introducing ML tools.…”

“…16−18 More importantly, ML has shown a strong ability to accelerate the discovery of new materials by effectively learning from a large amount of data, 19,20 but so far there are few studies regarding the ML assisted exploration of nanozymes. 21,22 Hence, it is highly desirable to accelerate the discovery of SOD mimics by introducing ML tools.…”