AutoGenome: An AutoML tool for genomic research

Deep learning has achieved great successes in traditional fields like computer vision (CV), natural language processing (NLP), speech processing, and more. These advancements have greatly inspired researchers genomics made deep an exciting popular topic. The convolutional neural network (CNN) recurrent (RNN) are frequently used to solve genomic sequencing prediction problems, multiple layer perception (MLP) auto-encoders (AE) for profiling data RNA expression gene mutation data. Here, we introduce a new architecture-the residual fully-connected (RFCN)-and describe its advantage modeling We also incorporate AutoML algorithms implement AutoGenome, end-to-end, automated framework studies. By utilizing the proposed RFCN architecture, automatic hyper-parameter search, architecture search algorithms, AutoGenome can automatically train high-performance models various kinds of To help better understand trained models, assess importance different features export most critical supervised tasks representative latent vectors unsupervised tasks. expect will become tool