Artificial image objects for classification of schizophrenia with GWAS-selected SNVs and convolutional neural network

In this article, we propose a new approach to analyze large genomics data. We considered individual genetic variants as pixels in an image and transformed a collection of variants into an artificial image object (AIO), which could be classified as a regular image by CNN algorithms. Using schizophrenia as a case study, we demonstrate the principles and their applications with 3 datasets. With 4,096 SNVs, the CNN models achieved an accuracy of 0.678 ± 0.007 and an AUC of 0.738 ± 0.008 for the diagnosis phenotype. With 44,100 SNVs, the models achieved class-specific accuracies of 0.806 ± 0.032 and 0.820 ± 0.049, and AUCs of 0.930 ± 0.017 and 0.867 ± 0.040 for the bottom and top classes stratified by the patient's polygenic risk scores. These results suggest that, once transformed to images, large genomics data can be analyzed effectively with image classification algorithms.

•

Introduce a technique to transform genomics data into AIOs

•

Apply CNN algorithms to classify genomics derived AIOs

•

Showcase the technique with GWAS-selected SNVs to classify schizophrenia diagnosis

Genome-wide association studies have discovered many genetic variants that contribute to human diseases. However, it remains a challenge to effectively utilize these variants to facilitate early and accurate diagnosis and treatment. In this report, we propose a new approach that transforms genetic data into AIOs so that they can be classified by advanced artificial intelligence and machine learning algorithms. Using schizophrenia as a case study, we demonstrate that genetic variants can be transformed into AIOs and that the AIOs can be classified by CNN algorithms consistently. Our approach can be applied to other omics data and combine them to jointly model disease risks and treatment responses.