Interpretable detection of novel human viruses from genome sequencing data

Abstract

Viruses evolve extremely quickly, so reliable methods for viral host prediction are necessary to safeguard biosecurity and biosafety alike. Novel human-infecting viruses are difficult to detect with standard bioinformatics workflows. Here, we predict whether a virus can infect humans directly from next-generation sequencing reads. We show that deep neural architectures significantly outperform both shallow machine learning and standard, homology-based algorithms, cutting the error rates in half and generalizing to taxonomic units distant from those presented during training. Further, we develop a suite of interpretability tools and show that it can be applied also to other models beyond the host prediction task. We propose a new approach for convolutional filter visualization to disentangle the information content of each nucleotide from its contribution to the final classification decision. Nucleotide-resolution maps of the learned associations between pathogen genomes and the infectious phenotype can be used to detect regions of interest in novel agents, for example, the SARS-CoV-2 coronavirus, unknown before it caused a COVID-19 pandemic in 2020. All methods presented here are implemented as easy-to-install packages not only enabling analysis of NGS datasets without requiring any deep learning skills, but also allowing advanced users to easily train and explain new models for genomics.

Deep learning-based real-time detection of novel pathogens during sequencing

“…Throughout this paper, we will use the term subread in a special sense: the first k nucleotides of a given sequencing read (in other words, a prefix of a read). The original DeePaC and DeePaC-vir datasets [23, 24] consist of 250bp simulated Illumina reads. The training, validation and held-out test sets contain mixtures of reads originating from different viruses or bacterial species with confirmed labels [32, 33], explicitly modeling generalization to “novel” (i.e.…”

“…We investigated two architectures shown previously to perform well in the pathogenicity or host-range prediction task – a reverse-complement CNN consisting of 2 convolutional layers and 2 fully-connected layers and a reverse-complement bidirectional LSTM. For more design details and the description of the reverse-complement variants of convolutional and LSTM layers, we refer the reader to [23, 24]. Those architectures guarantee identical predictions for sequences in their forward and reverse-complement orientations in a single forward pass.…”

“…Zhang et al [22] used a kNN classifier to develop a similar method for detection of human-infecting viruses. An analogous deep learning approach, DeePaC, outperforms the traditional machine learning algorithms on both novel bacteria [23] and viruses (DeePaC-vir [24]), offering an additional level of interpretability on nucleotide, read and genome levels. A similar method [25] focuses on detailed predictions for a small set of three viral species and cannot be used in an open-view setting.…”

“…While pathogenicity prediction methods using contigs, whole genomes or protein sets as input also exist, this work focuses on read-based classification to offer real-time predictions and avoid delays necessitated by assembly pipelines. However, read-based methods have been shown to perform well and also for full genomes and assembled contigs, achieving similar or better performance than alignment-based approaches [14, 23, 24]. This mirrors successful adoption of CNNs using raw DNA sequences as inputs in other fields ranging from regulatory genomics [27, 28, 29] to viral bioinformatics [25, 26, 30] and biosecurity [31].…”

Predicting the animal hosts of coronaviruses from compositional biases of spike protein and whole genome sequences through machine learning

“…These methods are capable of decomposing signal in high-dimensional genomic information (a limitation of regression frameworks) without the need for sequence alignment. Genomic machine learning analyses have demonstrated the ability to not only classify viruses from recurring viral genome motifs [28], but also classify their broad host origins [29][30][31][32]. Specifically considering coronaviruses, support vector machines and random forests have been trained on various genomic features to predict host group, including nucleotide and dinucleotide biases [33], amino acid composition [34] or sequence k-mers [35].…”