Florian Mock

Taxonomic classification, that is, the assignment to biological clades with shared ancestry, is a common task in genetics, mainly based on a genome similarity search of large genome databases. The classification quality depends heavily on the database, since representative relatives must be present. Many genomic sequences cannot be classified at all or only with a high misclassification rate. Here we present BERTax, a deep neural network program based on natural language processing to precisely classify the superkingdom and phylum of DNA sequences taxonomically without the need for a known representative relative from a database. We show BERTax to be at least on par with the state-of-the-art approaches when taxonomically similar species are part of the training data. For novel organisms, however, BERTax clearly outperforms any existing approach. Finally, we show that BERTax can also be combined with database approaches to further increase the prediction quality in almost all cases. Since BERTax is not based on similar entries in databases, it allows precise taxonomic classification of a broader range of genomic sequences, thus increasing the overall information gain.

Taxonomic classification, i.e., the identification and assignment to groups of biological organisms with the same origin and characteristics, is a common task in genetics. Nowadays, taxonomic classification is mainly based on genome similarity search to large genome databases. In this process, the classification quality depends heavily on the database since representative relatives have to be known already. Many genomic sequences cannot be classified at all or only with a high misclassification rate. Here we present BERTax , a program that uses a deep neural network to pre-cisely classify the superkingdom, phylum, and genus of DNA sequences taxonomically without the need for a known representative relative from a database. For this, BERTax uses the natural language processing model BERT trained to represent DNA. We show BERTax to be at least on par with the state-of-the-art approaches when taxonomically similar species are part of the training data. In case of an entirely novel organism, however, BERTax clearly outperforms any existing approach. Finally, we show that BERTax can also be combined with database approaches to further increase the prediction quality. Since BERTax is not based on homologous entries in databases, it allows precise taxonomic classification of a broader range of genomic sequences. This leads to a higher number of correctly classified sequences and thus increases the overall information gain.

Motivation: By binding to specific structures on antigenic proteins, the so-called epitopes, B-cell antibodies can neutralize pathogens. The identification of B-cell epitopes is of great value for the development of specific serodiagnostic assays and the optimization of medical therapy. However, identifying diagnostically or therapeutically relevant epitopes is a challenging task that usually involves extensive laboratory work. In this study, we show that the time, cost and labor-intensive process of epitope detection in the lab can be significantly reduced using in silico prediction. Results: Here, we present EpiDope, a python tool which uses a deep neural network to detect linear B-cell epitope regions on individual protein sequences. With an area under the curve between 0.67 ± 0.07 in the receiver operating characteristic curve, EpiDope exceeds all other currently used linear B-cell epitope prediction tools. Our software is shown to reliably predict linear B-cell epitopes of a given protein sequence, thus contributing to a significant reduction of laboratory experiments and costs required for the conventional approach. Availabilityand implementation: EpiDope is available on GitHub (http://github.com/mcollatz/EpiDope). Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: Zoonosis, the natural transmission of infections from animals to humans, is a far-reaching global problem. The recent outbreaks of Zikavirus, Ebolavirus and Coronavirus are examples of viral zoonosis, which occur more frequently due to globalization. In case of a virus outbreak, it is helpful to know which host organism was the original carrier of the virus to prevent further spreading of viral infection. Recent approaches aim to predict a viral host based on the viral genome, often in combination with the potential host genome and arbitrarily selected features. These methods are limited in the number of different hosts they can predict or the accuracy of the prediction. Results: Here, we present a fast and accurate deep learning approach for viral host prediction, which is based on the viral genome sequence only. We tested our deep neural network (DNN) on three different virus species (influenza A virus, rabies lyssavirus and rotavirus A). We achieved for each virus species an AUC between 0.93 and 0.98, allowing highly accurate predictions while using only fractions (100–400 bp) of the viral genome sequences. We show that deep neural networks are suitable to predict the host of a virus, even with a limited amount of sequences and highly unbalanced available data. The trained DNNs are the core of our virus–host prediction tool VIrus Deep learning HOst Prediction (VIDHOP). VIDHOP also allows the user to train and use models for other viruses. Availability and implementation: VIDHOP is freely available under https://github.com/flomock/vidhop. Supplementary information: Supplementary data are available at Bioinformatics online.

By binding to specific structures on antigenic proteins, the so-called epitopes, B-cell antibodies can neutralize pathogens. The identification of B-cell epitopes is of great value for the development of specific serodiagnostic assays and the optimization of medical therapy. However, identifying diagnostically or therapeutically relevant epitopes is a challenging task that usually involves extensive laboratory work. In this study, we show that the time, cost and labor-intensive process of epitope detection in the lab can be significantly shortened by using in silico prediction. Here we present EpiDope , a python tool which uses a deep neural network to detect B-cell epitope regions on individual protein sequences ( github.com/mcollatz/EpiDope ). With an area under the curve (AUC) between 0.67 ± 0.07 in the ROC curve, EpiDope exceeds all other currently used B-cell prediction tools. Moreover, for AUC10% (AUC for a false-positive rate < 0.1), EpiDope improves the prediction accuracy in comparison to other state-of-the-art methods. Our software is shown to reliably predict linear B-cell epitopes of a given protein sequence, thus contributing to a significant reduction of laboratory experiments and costs required for the conventional approach.

Although bats are increasingly becoming the focus of scientific studies due to their unique properties, these exceptional animals are still among the least studied mammals. Assembly quality and completeness of bat genomes vary a lot and especially non-coding RNA (ncRNA) annotations are incomplete or simply missing. Accordingly, standard bioinformatics pipelines for gene expression analysis often ignore ncRNAs such as microRNAs or long antisense RNAs. The main cause of this problem is the use of incomplete genome annotations. We present a complete screening for ncRNAs within 16 bat genomes. NcRNAs affect a remarkable variety of vital biological functions, including gene expression regulation, RNA processing, RNA interference and, as recently described, regulatory processes in viral infections. Within all investigated bat assemblies, we annotated 667 ncRNA families including 162 snoRNAs and 193 miRNAs as well as rRNAs, tRNAs, several snRNAs and lncRNAs, and other structural ncRNA elements. We validated our ncRNA candidates by six RNA-Seq data sets and show significant expression patterns that have never been described before in a bat species on such a large scale. Our annotations will be usable as a resource (rna.uni-jena.de/supplements/bats) for deeper studying of bat evolution, ncRNAs repertoire, gene expression and regulation, ecology and important host–virus interactions.

Although bats are increasingly becoming the focus of scientific studies due to their unique properties, these exceptional animals are still among the least studied mammals. Assembly quality and completeness of bat genomes vary a lot and especially non-coding RNA (ncRNA) annotations are incomplete or simply missing. Accordingly, standard bioinformatics pipelines for gene expression analysis often ignore ncRNAs such as microRNAs or long antisense RNAs. The main cause of this problem is the use of incomplete genome annotations. We present a complete screening for ncRNAs within 16 bat genomes. NcRNAs affect a remarkable variety of vital biological functions, including gene expression regulation, RNA processing, RNA interference and, as recently described, regulatory processes in viral infections. Within all investigated bat assemblies we annotated 667 ncRNA families including 162 snoRNAs and 193 miRNAs as well as rRNAs, tRNAs, several snRNAs and IncRNAs, and other structural ncRNA elements. We validated our ncRNA candidates by six RNA-Seq data sets and show significant expression patterns that have never been described before in a bat species on such a large scale. Our annotations will be usable as a resource (Electronic Supplement) for deeper studying of bat evolution, ncRNAs repertoire, gene expression and regulation, ecology, and important host-virus interactions. Supplementary information: is available at rna.uni-jena.de/supplements/bats , the Open Science Framework ( doi.org/10.17605/OSF.IO/4CMDN ), and GitHub ( github.com/rnajena/bats_ncrna ).

Motivation: Zoonosis, the natural transmission of infections from animals to humans, is a far-reaching global problem. The recent outbreaks of Zika virus, Ebola virus and Corona virus are examples of viral zoonosis, which occur more frequently due to globalization. In the case of a virus outbreak, it is helpful to know which host organism was the original carrier of the virus. Once the reservoir or intermediate host is known, it can be isolated to prevent further spreading of the viral infection. Recent approaches aim to predict a viral host based on the viral genome, often in combination with the potential host genome and arbitrarily selected features. These methods have a clear limitation in either the number of different hosts they can predict or the accuracy of their prediction. Results: Here, we present a fast and accurate deep learning approach for viral host prediction, which is based on the viral genome sequence only. To ensure a high prediction accuracy, we developed an effective selection approach for the training data to avoid biases due to a highly unbalanced number of known sequences per virus-host combinations. We tested our deep neural network on three different virus species (influenza A, rabies lyssavirus, rotavirus A). We reached for each virus species an AUG between 0.93 and 0.98, outperforming previous approaches and allowing highly accurate predictions while only using fractions (100-400 bp) of the viral genome sequences. We show that deep neural networks are suitable to predict the host of a virus, even with a limited amount of sequences and highly unbalanced available data. The deep neural networks trained for this approach build the core of the virus-host predicting tool VIDHOP (Virus Deep learning HOst Prediction). Availability: The trained models for the prediction of the host for the viruses influenza A, rabies lyssavirus, rotavirus A are implemented in the tool VIDHOP. This tool is freely available under https://github.com/flomock/vidhop . Supplementary information: Supplementary data are available at DOI 10.17605/OSF.IO/UXT7N