A k-mer-based barcode DNA classification methodology based on spectral representation and a neural gas network

A Graph Neural Network Approach for the Analysis of siRNA-Target Biological Networks

Many biological systems are characterised by biological entities, as well as their relationships. These interaction networks can be modelled as graphs, with nodes representing bio-entities, such as molecules, and edges representing relations among them, such as interactions. Due to the current availability of a huge amount of biological data, it is very important to consider in silico analysis methods based on, for example, machine learning, that could take advantage of the inner graph structure of the data in order to improve the quality of the results. In this scenario, graph neural networks (GNNs) are recent computational approaches that directly deal with graph-structured data. In this paper, we present a GNN network for the analysis of siRNA-mRNA interaction networks. siRNAs, in fact, are small RNA molecules that are able to bind to target genes and silence them. These events make siRNAs key molecules as RNA interference agents in many biological interaction networks related to severe diseases such as cancer. In particular, our GNN approach allows for the prediction of the siRNA efficacy, which measures the siRNA's ability to bind and silence a gene target. Tested on benchmark datasets, our proposed method overcomes other machine learning algorithms, including the state-of-the-art predictor based on the convolutional neural network, reaching a Pearson correlation coefficient of approximately 73.6%. Finally, we proposed a case study where the efficacy of a set of siRNAs is predicted for a gene of interest. To the best of our knowledge, GNNs were used for the first time in this scenario.

Mathematical Modeling and Computational Prediction of High-Risk Types of Human Papillomaviruses

Cervical cancer is one of the main causes of cancer death all over the world. Most diseases such as cervical epithelial atypical hyperplasia and invasive cervical cancer are closely related to the continuous infection of high-risk types of human papillomavirus. Therefore, the high-risk types of human papillomavirus are the key to the prevention and treatment of cervical cancer. With the accumulation of high-throughput and clinical data, the use of systematic and quantitative methods for mathematical modeling and computational prediction has become more and more important. This paper summarizes the mathematical models and prediction methods of the risk types of human papillomavirus, especially around the key steps such as feature extraction, feature selection, and prediction algorithms. We summarized and discussed the advantages and disadvantages of existing algorithms, which provides a theoretical basis for follow-up research.

DeepBarcoding: Deep Learning for Species Classification Using DNA Barcoding

DNA barcodes with short sequence fragments are used for species identification. Because of advances in sequencing technologies, DNA barcodes have gradually been emphasized. DNA sequences from different organisms are easily and rapidly acquired. Therefore, DNA sequence analysis tools play an increasingly crucial role in species identification. This study proposed deep barcoding, a deep learning framework for species classification by using DNA barcodes. Deep barcoding uses raw sequence data as the input to represent one-hot encoding as a one-dimensional image and uses a deep convolutional neural network with a fully connected deep neural network for sequence analysis. It can achieve an average accuracy of >90 percent for both simulation and real datasets. Although deep learning yields outstanding performance for species classification with DNA sequences, its application remains a challenge. The deep barcoding model can be a potential tool for species classification and can elucidate DNA barcode-based species identification.

Evaluation of Arabian Vascular Plant Barcodes (rbcL and matK): Precision of Unsupervised and Supervised Learning Methods towards Accurate Identification

Arabia is the largest peninsula in the world, with >3000 species of vascular plants. Not much effort has been made to generate a multi-locus marker barcode library to identify and discriminate the recorded plant species. This study aimed to determine the reliability of the available Arabian plant barcodes (>1500; rbcL and matK) at the public repository (NCBI GenBank) using the unsupervised and supervised methods. Comparative analysis was carried out with the standard dataset (FINBOL) to assess the methods and markers' reliability. Our analysis suggests that from the unsupervised method, TaxonDNA's All Species Barcode criterion (ASB) exhibits the highest accuracy for rbcL barcodes, followed by the matK barcodes using the aligned dataset (FINBOL). However, for the Arabian plant barcode dataset (GBMA), the supervised method performed better than the unsupervised method, where the Random Forest and K-Nearest Neighbor (gappy kernel) classifiers were robust enough. These classifiers successfully recognized true species from both barcode markers belonging to the aligned and alignment-free datasets, respectively. The multi-class classifier showed high species resolution following the two classifiers, though its performance declined when employed to recognize true species. Similar results were observed for the FINBOL dataset through the supervised learning approach; overall, matK marker showed higher accuracy than rbcL. However, the lower rate of species identification in matK in GBMA data could be due to the higher evolutionary rate or gaps and missing data, as observed for the ASB criterion in the FINBOL dataset. Further, a lower number of sequences and singletons could also affect the rate of species resolution, as observed in the GBMA dataset. The GBMA dataset lacks sufficient species membership. We would encourage the taxonomists from the Arabian Peninsula to join our campaign on the Arabian Barcode of Life at the Barcode of Life Data (BOLD) systems. Our efforts together could help improve the rate of species identification for the Arabian Vascular plants.

Accurate prediction of RNA 5-hydroxymethylcytosine modification by utilizing novel position-specific gapped k-mer descriptors

RNA modification is an essential step towards generation of new RNA structures. Such modification is potentially able to modify RNA function or its stability. Among different modifications, 5-Hydroxymethylcytosine (5hmC) modification of RNA exhibit significant potential for a series of biological processes. Understanding the distribution of 5hmC in RNA is essential to determine its biological functionality. Although conventional sequencing techniques allow broad identification of 5hmC, they are both time-consuming and resource-intensive. In this study, we propose a new computational tool called iRNA5hmC-PS to tackle this problem. To build iRNA5hmC-PS we extract a set of novel sequence-based features called Position-Specific Gapped k-mer (PSG k-mer) to obtain maximum sequential information. Our feature analysis shows that our proposed PSG k-mer features contain vital information for the identification of 5hmC sites. We also use a group-wise feature importance calculation strategy to select a small subset of features containing maximum discriminative information. Our experimental results demonstrate that iRNA5hmC-PS is able to enhance the prediction performance, dramatically. iRNA5hmC-PS achieves 78.3% prediction performance, which is 12.8% better than those reported in the previous studies. iRNA5hmC-PS is publicly available as an online tool at http://103.109.52.8:81/iRNA5hmC-PS. Its benchmark dataset, source codes, and documentation are available at https://github.com/zahid6454/iRNA5hmC-PS.

DeepTE: a computational method for de novo classification of transposons with convolutional neural network

Motivation

Transposable elements (TEs) classification is an essential step to decode their roles in genome evolution. With a large number of genomes from non-model species becoming available, accurate and efficient TE classification has emerged as a new challenge in genomic sequence analysis.

Results

We developed a novel tool, DeepTE, which classifies unknown TEs using convolutional neural networks (CNNs). DeepTE transferred sequences into input vectors based on k-mer counts. A tree structured classification process was used where eight models were trained to classify TEs into super families and orders. DeepTE also detected domains inside TEs to correct false classification. An additional model was trained to distinguish between non-TEs and TEs in plants. Given unclassified TEs of different species, DeepTE can classify TEs into seven orders, which include 15, 24 and 16 super families in plants, metazoans and fungi, respectively. In several benchmarking tests, DeepTE outperformed other existing tools for TE classification. In conclusion, DeepTE successfully leverages CNN for TE classification, and can be used to precisely classify TEs in newly sequenced eukaryotic genomes.

Availability and implementation

DeepTE is accessible at https://github.com/LiLabAtVT/DeepTE.

Supplementary information

Supplementary data are available at Bioinformatics online.

Methylation-driven model for analysis of dinucleotide evolution in genomes

Background

CpGs, the major methylation sites in vertebrate genomes, exhibit a high mutation rate from the methylated form of CpG to TpG/CpA and, therefore, influence the evolution of genome composition. However, the quantitative effects of CpG to TpG/CpA mutations on the evolution of genome composition in terms of the dinucleotide frequencies/proportions remain poorly understood.

Results

Based on the neutral theory of molecular evolution, we propose a methylation-driven model (MDM) that allows predicting the changes in frequencies/proportions of the 16 dinucleotides and in the GC content of a genome given the known number of CpG to TpG/CpA mutations. The application of MDM to the 10 published vertebrate genomes shows that, for most of the 16 dinucleotides and the GC content, a good consistency is achieved between the predicted and observed trends of changes in the frequencies and content relative to the assumed initial values, and that the model performs better on the mammalian genomes than it does on the lower-vertebrate genomes. The model’s performance depends on the genome composition characteristics, the assumed initial state of the genome, and the estimated parameters, one or more of which are responsible for the different application effects on the mammalian and lower-vertebrate genomes and for the large deviations of the predicted frequencies of a few dinucleotides from their observed frequencies.

Conclusions

Despite certain limitations of the current model, the successful application to the higher-vertebrate (mammalian) genomes witnesses its potential for facilitating studies aimed at understanding the role of methylation in driving the evolution of genome dinucleotide composition.

Deep learning architectures for prediction of nucleosome positioning from sequences data

Background

Nucleosomes are DNA-histone complex, each wrapping about 150 pairs of double-stranded DNA. Their function is fundamental for one of the primary functions of Chromatin i.e. packing the DNA into the nucleus of the Eukaryote cells. Several biological studies have shown that the nucleosome positioning influences the regulation of cell type-specific gene activities. Moreover, computational studies have shown evidence of sequence specificity concerning the DNA fragment wrapped into nucleosomes, clearly underlined by the organization of particular DNA substrings. As the main consequence, the identification of nucleosomes on a genomic scale has been successfully performed by computational methods using a sequence features representation.

Results

In this work, we propose a deep learning model for nucleosome identification. Our model stacks convolutional layers and Long Short-term Memories to automatically extract features from short- and long-range dependencies in a sequence. Using this model we are able to avoid the feature extraction and selection steps while improving the classification performances.

Conclusions

Results computed on eleven data sets of five different organisms, from Yeast to Human, show the superiority of the proposed method with respect to the state of the art recently presented in the literature.

Deep learning models for bacteria taxonomic classification of metagenomic data

Background

An open challenge in translational bioinformatics is the analysis of sequenced metagenomes from various environmental samples. Of course, several studies demonstrated the 16S ribosomal RNA could be considered as a barcode for bacteria classification at the genus level, but till now it is hard to identify the correct composition of metagenomic data from RNA-seq short-read data. 16S short-read data are generated using two next generation sequencing technologies, i.e. whole genome shotgun (WGS) and amplicon (AMP); typically, the former is filtered to obtain short-reads belonging to a 16S shotgun (SG), whereas the latter take into account only some specific 16S hypervariable regions. The above mentioned two sequencing technologies, SG and AMP, are used alternatively, for this reason in this work we propose a deep learning approach for taxonomic classification of metagenomic data, that can be employed for both of them.

Results

To test the proposed pipeline, we simulated both SG and AMP short-reads, from 1000 16S full-length sequences. Then, we adopted a k-mer representation to map sequences as vectors into a numerical space. Finally, we trained two different deep learning architecture, i.e., convolutional neural network (CNN) and deep belief network (DBN), obtaining a trained model for each taxon. We tested our proposed methodology to find the best parameters configuration, and we compared our results against the classification performances provided by a reference classifier for bacteria identification, known as RDP classifier. We outperformed the RDP classifier at each taxonomic level with both architectures. For instance, at the genus level, both CNN and DBN reached 91.3% of accuracy with AMP short-reads, whereas RDP classifier obtained 83.8% with the same data.

Conclusions

In this work, we proposed a 16S short-read sequences classification technique based on k-mer representation and deep learning architecture, in which each taxon (from phylum to genus) generates a classification model. Experimental results confirm the proposed pipeline as a valid approach for classifying bacteria sequences; for this reason, our approach could be integrated into the most common tools for metagenomic analysis. According to obtained results, it can be successfully used for classifying both SG and AMP data.

Electronic supplementary material

The online version of this article (10.1186/s12859-018-2182-6) contains supplementary material, which is available to authorized users.