Sparse Supervised Classification Methods Predict and Characterize Nanomaterial Exposures: Independent Markers of MWCNT Exposures

Abstract

Recent experimental evidence indicates significant pulmonary toxicity of multiwalled carbon nanotubes (MWCNTs), such as inflammation, interstitial fibrosis, granuloma formation, and carcinogenicity. Although numerous studies explored the adverse potential of various CNTs, their comparability is often limited. This is due to differences in administered dose, physicochemical characteristics, exposure methods, and end points monitored. Here, we addressed the problem through sparse classification method, a supervised machine learning approach that can reduce the noise contained in redundant variables for discriminating among MWCNT-exposed and MWCNT-unexposed groups. A panel of proteins measured from bronchoalveolar lavage fluid (BAL) samples was used to predict exposure to various MWCNT and determine markers that are attributable to MWCNT exposure and toxicity in mice. Using sparse support vector machine-based classification technique, we identified a small subset of proteins clearly distinguishing each exposure. Macrophage-derived chemokine (MDC/CCL22), in particular, was associated with various MWCNT exposures and was independent of exposure method employed, that is, oropharyngeal aspiration versus inhalation exposure. Sustained expression of some of the selected protein markers identified also suggests their potential role in MWCNT-induced toxicity and proposes hypotheses for future mechanistic studies. Such approaches can be used more broadly for nanomaterial risk profiling studies to evaluate decisions related to dose/time-response relationships that could delineate experimental variables from exposure markers.

Physicochemical characterization and genotoxicity of the broad class of carbon nanotubes and nanofibers used or produced in U.S. facilities

“…Advances in computational analysis are being applied to the almost two decades of engineered nanomaterial research for grouping and understanding the physicochemical drivers of toxicity [95-97, 119, 120], including studies of carbon nanotubes [45,50,77,121,122]. The analyses of the data from this study illustrate that detailed physical dimension characteristics provide a more consistent grouping of CNT/F as compared to using only data means.…”

Practices and Trends of Machine Learning Application in Nanotoxicology

“…NMs Category Output Reference NMs Category Output [118] Carbon-based exposed/not exposed groups [60] Metal Oxide Viability [76] Metal The kNN method classifies a case in the feature space based on the nearest training instances [62] relying on the similarity principle [40]. Based on weighted majority voting, each case is allocated to the class of the kth closest neighbors.…”

“…The sparse classification Lone-Star algorithm implements optimization methods to overcome issues inherent to nanotoxicity modeling, such as unequal distribution of classes and unknown relationships between inputs. This method, when compared to traditional SVMs, takes advantage of the combined l1-norm and l2-norm SVM's ability to select a small set of features while ignoring the redundant ones to achieve both the classification goal and the selection of correlated features simultaneously [118].…”

“…MCC is equivalent to the Pearson Correlation Coefficient for binary variables [147] and has been selected as an evaluation metric for micro-array-based predictive models by the MicroArray Quality Control (MAQC) Consortium [148]. From the gathered studies, only six used MCC as a performance metric [77,118,121,122,124,126]. Table 10 presents the studies that, in compliance with OECD principles, applied measures of robustness and predictivity validation, and estimated the applicability domain.…”

Data Science Applied to Carbon Materials: Synthesis, Characterization, and Applications

“…[ 1 ] Carbon materials in conjunction with data science tools have been used to improve viral detection and study carbon material toxicology. [ 21,64,65,83 ] Viruses can become a great threat to health and society, as observed with the recent outbreak of COVID‐19, with great socioeconomic implications. Therefore, fast and reliable methods for virus detection are necessary.…”