Cell Type-Specific Predictive Models Perform Prioritization of Genes and Gene Sets Associated With Autism

Abstract

Bulk transcriptomic analyses of autism spectrum disorder (ASD) have revealed dysregulated pathways, while the brain cell type-specific molecular pathology of ASD still needs to be studied. Machine learning-based studies can be conducted for ASD, prioritizing high-confidence gene candidates and promoting the design of effective interventions. Using human brain nucleus gene expression of ASD and controls, we construct cell type-specific predictive models for ASD based on individual genes and gene sets, respectively, to screen cell type-specific ASD-associated genes and gene sets. These two kinds of predictive models can predict the diagnosis of a nucleus with known cell type. Then, we construct a multi-label predictive model for predicting the cell type and diagnosis of a nucleus at the same time. Our findings suggest that layer 2/3 and layer 4 excitatory neurons, layer 5/6 cortico-cortical projection neurons, parvalbumin interneurons, and protoplasmic astrocytes are preferentially affected in ASD. The functions of genes with predictive power for ASD are different and the top important genes are distinct across different cells, highlighting the cell-type heterogeneity of ASD. The constructed predictive models can promote the diagnosis of ASD, and the prioritized cell type-specific ASD-associated genes and gene sets may be used as potential biomarkers of ASD.

Neuroimaging genetics approaches to identify new biomarkers for the early diagnosis of autism spectrum disorder

“…Moreover, the efficacy of DL methods can be enhanced by implementing predictive modeling and collaborative filtering techniques. Predictive modeling is a statistical technique that uses ML and data mining to forecast or predict future outcomes and can help promote the diagnosis of ASD [ 216 , 217 ]. Collaborative filtering is another technique that can allow the creation of personalized recommendations based on the interactions, data, and preferences of other users in the system [ 218 ].…”

Single-nucleus gene and gene set expression-based similarity network fusion identifies autism molecular subtypes

“…Constructing machine learning-based disease prediction model can facilitate the clinical diagnosis of diseases [ 41 – 43 ]. Using the identified ASD molecular subtype-specific DEGs, we built PLS-based prediction models for classifying molecular subtypes.…”

HEC-ASD: a hybrid ensemble-based classification model for predicting autism spectrum disorder disease genes

“…The first part used the hybrid gene similarity function (HGS) with Ada-Boost ensemble learning machines. Figure 11 shows a detailed performance measure in terms of true positive rate (TP Rate), false positive rate (FP Rate), Precision, Recall, (11) Accuracy = TP + TN TP + FP + TN + FN F-measure, AUC-ROC and Accuracy, which reached 84.35%, increasing the accuracy of Random Forest by around 4.5%. Moreover, HEC-ASD based on a gradient boosting model used regularization parameters to prevent overfitting the model.…”

“…Machine learning techniques are helpful in different biomedical problems such as drug discovery [ 2 , 3 ], gene prediction [ 4 , 5 ], disease gene interactions [ 6 ], genome annotations [ 7 ], gene regulatory network derivation [ 8 ], microarray data classification [ 9 ],and protein function prediction [ 10 ]. Some predictive models [ 11 ] were proposed to identify genes related to ASD and gene sets associated with autism according to specific cell types. They give higher priority to high-confidence candidates genes from Simons Foundation Autism Research Initiative (SFARI) to construct a predictive model.…”