GenotypeTensors: Efficient Neural Network Genotype Callers

Abstract

We studied the problem of calling genotypes using neural networks. A machine learning approach to calling genotypes requires a training set, an approach to convert genomic sites into tensors and robust model development and evaluation protocols. We discuss each of these components of our approach and compare four types of neural network training protocols, two fully supervised and two semi-supervised approaches. Semi-supervised approaches use unlabeled data to supplement limited quantities of labeled data. Random hyper-parameter searches identified highly performing models that reach indel F1 of 99.4% on a chromosomes 20, 21, 22 and X of NA12878/HG001. We further validate these models by evaluating performance on HG002, an independent sample used in the PrecisionFDA challenge. We apply GenotypeTensors to evaluate the impact of (1) training with small datasets, (2) training models only with sites inside confidence regions, or (3) training with improved true label annotations. A PyTorch open-source implementation of GenotypeTensors is available at https://github.com/CampagneLaboratory/GenotypeTensors. DNANexus cloud applications are provided to help process new datasets both to train model or call genotypes with trained models.Keywords: Deep Learning, Machine Learning, Genotype Caller, High-Throughput Sequencing Recent work showed that careful tuning of baseline architectures can yield state of the art performance compared to more complex architectures [Merity et al., 2018] (authors studied sequence models for natural language processing tasks). This study confirms that hyper-parameter tuning is critical to training state of the art neural network models. In practice, selecting optimal hyper-parameters is difficult because of the computational burden of training many models with different hyper-parameters. In this study, we present and take advantage of an approach that greatly speeds up hyper-parameter searches when the models are small and many models can fit in the memory of a single graphical processing unit (GPU). Figure 1 presents an overview of the process we followed to prepare data for neural network training. Briefly, short reads were aligned to the human genome, alignments were processed with HaplotypeCaller [McKenna et al., 2010] to realign SNPs in the proximity of indels and to reduce the dimensionality of the dataset to regions likely to contain variation. Alignments were converted to a vectorial representation suitable to train a feed-forward neural network. Figure 1 also illustrates the funnel architecture, which allows for interactions of every feature with every other feature and progressively 2/13 RESULTS Data Preparation

HetEnc: A Deep Learning Predictive Model for Multi-type Biological Dataset

“…Deep learning, one of the most promising methods in current machine learning, has been implemented in a variety of research fields, including object recognition, keyword triggering, language translation, and others [10]. It has been applied in such biological study areas as variant calling [11,12], protein-binding prediction [13], predicting variant chromatin effects [14], and biomedical imaging [15][16][17]. Note that most of these biological applications were applied to spatial/temporal/sequential data, for which many deep learning approaches have been developed in other research fields.…”

HetEnc: A Deep Learning Predictive Model for Multi-type Biological Dataset

“…Deep learning, one of the most promising methods in current machine learning, has been implemented in a variety of research fields, including object recognition, keyword triggering, language translation, and others [7]. It has been applied in such biological study areas as variant calling [8,9], protein-binding prediction [10], predicting variant chromatin effects [11], and biomedical imaging [12][13][14]. Note that most of these biological applications were applied to spatial/temporal/sequential data, for which many deep learning approaches have been developed in other research fields.…”

A survey of dimension reduction and classification methods for RNA-Seq data on malaria vector

“…Currently machine learning technology has one of the greatest and promising approaches, they have been proposed and applied in a diverse of research works, such as recognition, language translation, bioinformatics and others [55]. RNA-seq skills are speedily developing, and its major challenges is addressing the huge amount of procedural noises that can determine about half percentage of cell disparities in expression sizes [56].…”