Search

Proteins and nucleic-acids are essential components of living organisms that interact in critical cellular processes. Accurate prediction of nucleic acid-binding residues in proteins can contribute to a better understanding of protein function. However, the discrepancy between protein sequence information and obtained structural and functional data renders most current computational models ineffective. Therefore, it is vital to design computational models based on protein sequence information to identify nucleic acid binding sites in proteins. Here, we implement an ensemble deep learning model-based nucleic-acid-binding residues on proteins identification method, called SOFB, which characterizes protein sequences by learning the semantics of biological dynamics contexts, and then develop an ensemble deep learning-based sequence network to learn feature representation and classification by explicitly modeling dynamic semantic information. Among them, the language learning model, which is constructed from natural language to biological language, captures the underlying relationships of protein sequences, and the ensemble deep learning-based sequence network consisting of different convolutional layers together with Bi-LSTM refines various features for optimal performance. Meanwhile, to address the imbalanced issue, we adopt ensemble learning to train multiple models and then incorporate them. Our experimental results on several DNA/RNA nucleic-acid-binding residue datasets demonstrate that our proposed model outperforms other state-of-the-art methods. In addition, we conduct an interpretability analysis of the identified nucleic acid binding residue sequences based on the attention weights of the language learning model, revealing novel insights into the dynamic semantic information that supports the identified nucleic acid binding residues. SOFB is available at https://github.com/Encryptional/SOFB and https://figshare.com/articles/online_resource/SOFB_figshare_rar/25499452 ., (© 2024. The Author(s).)

Drug repurposing is promising in multiple scenarios, such as emerging viral outbreak controls and cost reductions of drug discovery. Traditional graph-based drug repurposing methods are limited to fast, large-scale virtual screens, as they constrain the counts for drugs and targets and fail to predict novel viruses or drugs. Moreover, though deep learning has been proposed for drug repurposing, only a few methods have been used, including a group of pre-trained deep learning models for embedding generation and transfer learning. Hence, we propose DeepSeq2Drug to tackle the shortcomings of previous methods. We leverage multi-modal embeddings and an ensemble strategy to complement the numbers of drugs and viruses and to guarantee the novel prediction. This framework (including the expanded version) involves four modal types: six NLP models, four CV models, four graph models, and two sequence models. In detail, we first make a pipeline and calculate the predictive performance of each pair of viral and drug embeddings. Then, we select the best embedding pairs and apply an ensemble strategy to conduct anti-viral drug repurposing. To validate the effect of the proposed ensemble model, a monkeypox virus (MPV) case study is conducted to reflect the potential predictive capability. This framework could be a benchmark method for further pre-trained deep learning optimization and anti-viral drug repurposing tasks. We also build software further to make the proposed model easier to reuse. The code and software are freely available at http://deepseq2drug.cs.cityu.edu.hk., Competing Interests: Declaration of competing interest None Declared, (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Single-cell RNA sequencing (scRNA-seq) is a robust method for studying gene expression at the single-cell level, but accurately quantifying genetic material is often hindered by limited mRNA capture, resulting in many missing expression values. Existing imputation methods rely on strict data assumptions, limiting their broader application, and lack reliable supervision, leading to biased signal recovery. To address these challenges, authors developed Bis, a distribution-agnostic deep learning model for accurately recovering missing sing-cell gene expression from multiple platforms. Bis is an optimal transport-based autoencoder model that can capture the intricate distribution of scRNA-seq data while addressing the characteristic sparsity by regularizing the cellular embedding space. Additionally, they propose a module using bulk RNA-seq data to guide reconstruction and ensure expression consistency. Experimental results show Bis outperforms other models across simulated and real datasets, showcasing superiority in various downstream analyses including batch effect removal, clustering, differential expression analysis, and trajectory inference. Moreover, Bis successfully restores gene expression levels in rare cell subsets in a tumor-matched peripheral blood dataset, revealing developmental characteristics of cytokine-induced natural killer cells within a head and neck squamous cell carcinoma microenvironment., (© 2024 The Authors. Advanced Science published by Wiley‐VCH GmbH.)

Motivation: Recent frameworks based on deep learning have been developed to identify cancer subtypes from high-throughput gene expression profiles. Unfortunately, the performance of deep learning is highly dependent on its neural network architectures which are often hand-crafted with expertise in deep neural networks, meanwhile, the optimization and adjustment of the network are usually costly and time consuming., Results: To address such limitations, we proposed a fully automated deep neural architecture search model for diagnosing consensus molecular subtypes from gene expression data (DNAS). The proposed model uses ant colony algorithm, one of the heuristic swarm intelligence algorithms, to search and optimize neural network architecture, and it can automatically find the optimal deep learning model architecture for cancer diagnosis in its search space. We validated DNAS on eight colorectal cancer datasets, achieving the average accuracy of 95.48%, the average specificity of 98.07%, and the average sensitivity of 96.24%, respectively. Without the loss of generality, we investigated the general applicability of DNAS further on other cancer types from different platforms including lung cancer and breast cancer, and DNAS achieved an area under the curve of 95% and 96%, respectively. In addition, we conducted gene ontology enrichment and pathological analysis to reveal interesting insights into cancer subtype identification and characterization across multiple cancer types., Availability and Implementation: The source code and data can be downloaded from https://github.com/userd113/DNAS-main. And the web server of DNAS is publicly accessible at 119.45.145.120:5001., (© The Author(s) 2023. Published by Oxford University Press.)

Protein-protein interactions (PPIs) govern cellular pathways and processes, by significantly influencing the functional expression of proteins. Therefore, accurate identification of protein-protein interaction binding sites has become a key step in the functional analysis of proteins. However, since most computational methods are designed based on biological features, there are no available protein language models to directly encode amino acid sequences into distributed vector representations to model their characteristics for protein-protein binding events. Moreover, the number of experimentally detected protein interaction sites is much smaller than that of protein-protein interactions or protein sites in protein complexes, resulting in unbalanced data sets that leave room for improvement in their performance. To address these problems, we develop an ensemble deep learning model (EDLM)-based protein-protein interaction (PPI) site identification method (EDLMPPI). Evaluation results show that EDLMPPI outperforms state-of-the-art techniques including several PPI site prediction models on three widely-used benchmark datasets including Dset_448, Dset_72, and Dset_164, which demonstrated that EDLMPPI is superior to those PPI site prediction models by nearly 10% in terms of average precision. In addition, the biological and interpretable analyses provide new insights into protein binding site identification and characterization mechanisms from different perspectives. The EDLMPPI webserver is available at http://www.edlmppi.top:5002/ ., (© 2023. The Author(s).)

The cooperativity of transcription factors (TFs) is a widespread phenomenon in the gene regulation system. However, the interaction patterns between TF binding motifs remain elusive. The recent high-throughput assays, CAP-SELEX, have identified over 600 composite DNA sites (i.e. heterodimeric motifs) bound by cooperative TF pairs. However, there are over 25 000 inferentially effective heterodimeric TFs in the human cells. It is not practically feasible to validate all heterodimeric motifs due to cost and labor. We introduce DeepMotifSyn, a deep learning-based tool for synthesizing heterodimeric motifs from monomeric motif pairs. Specifically, DeepMotifSyn is composed of heterodimeric motif generator and evaluator. The generator is a U-Net-based neural network that can synthesize heterodimeric motifs from aligned motif pairs. The evaluator is a machine learning-based model that can score the generated heterodimeric motif candidates based on the motif sequence features. Systematic evaluations on CAP-SELEX data illustrate that DeepMotifSyn significantly outperforms the current state-of-the-art predictors. In addition, DeepMotifSyn can synthesize multiple heterodimeric motifs with different orientation and spacing settings. Such a feature can address the shortcomings of previous models. We believe DeepMotifSyn is a more practical and reliable model than current predictors on heterodimeric motif synthesis. Contact:kc.w@cityu.edu.hk., (© The authors 2021. Published by Oxford University Press on behalf of the Institute of Mathematics and its Applications. All rights reserved.)

Single-cell RNA sequencing (scRNA-seq) technologies have been heavily developed to probe gene expression profiles at single-cell resolution. Deep imputation methods have been proposed to address the related computational challenges (e.g. the gene sparsity in single-cell data). In particular, the neural architectures of those deep imputation models have been proven to be critical for performance. However, deep imputation architectures are difficult to design and tune for those without rich knowledge of deep neural networks and scRNA-seq. Therefore, Surrogate-assisted Evolutionary Deep Imputation Model (SEDIM) is proposed to automatically design the architectures of deep neural networks for imputing gene expression levels in scRNA-seq data without any manual tuning. Moreover, the proposed SEDIM constructs an offline surrogate model, which can accelerate the computational efficiency of the architectural search. Comprehensive studies show that SEDIM significantly improves the imputation and clustering performance compared with other benchmark methods. In addition, we also extensively explore the performance of SEDIM in other contexts and platforms including mass cytometry and metabolic profiling in a comprehensive manner. Marker gene detection, gene ontology enrichment and pathological analysis are conducted to provide novel insights into cell-type identification and the underlying mechanisms. The source code is available at https://github.com/li-shaochuan/SEDIM., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Mitochondria are membrane-bound organelles containing over 1000 different proteins involved in mitochondrial function, gene expression and metabolic processes. Accurate localization of those proteins in the mitochondrial compartments is critical to their operation. A few computational methods have been developed for predicting submitochondrial localization from the protein sequences. Unfortunately, most of these computational methods focus on employing biological features or evolutionary information to extract sequence features, which greatly limits the performance of subsequent identification. Moreover, the efficiency of most computational models is still under explored, especially the deep learning feature, which is promising but requires improvement. To address these limitations, we propose a novel computational method called iDeepSubMito to predict the location of mitochondrial proteins to the submitochondrial compartments. First, we adopted a coding scheme using the ProteinELMo to model the probability distribution over the protein sequences and then represent the protein sequences as continuous vectors. Then, we proposed and implemented convolutional neural network architecture based on the bidirectional LSTM with self-attention mechanism, to effectively explore the contextual information and protein sequence semantic features. To demonstrate the effectiveness of our proposed iDeepSubMito, we performed cross-validation on two datasets containing 424 proteins and 570 proteins respectively, and consisting of four different mitochondrial compartments (matrix, inner membrane, outer membrane and intermembrane regions). Experimental results revealed that our method outperformed other computational methods. In addition, we tested iDeepSubMito on the M187, M983 and MitoCarta3.0 to further verify the efficiency of our method. Finally, the motif analysis and the interpretability analysis were conducted to reveal novel insights into subcellular biological functions of mitochondrial proteins. iDeepSubMito source code is available on GitHub at https://github.com/houzl3416/iDeepSubMito., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Haploinsufficiency, wherein a single allele is not enough to maintain normal functions, can lead to many diseases including cancers and neurodevelopmental disorders. Recently, computational methods for identifying haploinsufficiency have been developed. However, most of those computational methods suffer from study bias, experimental noise and instability, resulting in unsatisfactory identification of haploinsufficient genes. To address those challenges, we propose a deep forest model, called HaForest, to identify haploinsufficient genes. The multiscale scanning is proposed to extract local contextual representations from input features under Linear Discriminant Analysis. After that, the cascade forest structure is applied to obtain the concatenated features directly by integrating decision-tree-based forests. Meanwhile, to exploit the complex dependency structure among haploinsufficient genes, the LightGBM library is embedded into HaForest to reveal the highly expressive features. To validate the effectiveness of our method, we compared it to several computational methods and four deep learning algorithms on five epigenomic data sets. The results reveal that HaForest achieves superior performance over the other algorithms, demonstrating its unique and complementary performance in identifying haploinsufficient genes. The standalone tool is available at https://github.com/yangyn533/HaForest., (© The Author(s) 2021. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

The identification of hidden responders is often an essential challenge in precision oncology. A recent attempt based on machine learning has been proposed for classifying aberrant pathway activity from multiomic cancer data. However, we note several critical limitations there, such as high-dimensionality, data sparsity and model performance. Given the central importance and broad impact of precision oncology, we propose nature-inspired deep Ras activation pan-cancer (NatDRAP), a deep neural network (DNN) model, to address those restrictions for the identification of hidden responders. In this study, we develop the nature-inspired deep learning model that integrates bulk RNA sequencing, copy number and mutation data from PanCanAltas to detect pan-cancer Ras pathway activation. In NatDRAP, we propose to synergize the nature-inspired artificial bee colony algorithm with different gradient-based optimizers in one framework for optimizing DNNs in a collaborative manner. Multiple experiments were conducted on 33 different cancer types across PanCanAtlas. The experimental results demonstrate that the proposed NatDRAP can provide superior performance over other benchmark methods with strong robustness towards diagnosing RAS aberrant pathway activity across different cancer types. In addition, gene ontology enrichment and pathological analysis are conducted to reveal novel insights into the RAS aberrant pathway activity identification and characterization. NatDRAP is written in Python and available at https://github.com/lixt314/NatDRAP1., (© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Recent advances in high-throughput single-cell RNA-seq have enabled us to measure thousands of gene expression levels at single-cell resolution. However, the transcriptomic profiles are high-dimensional and sparse in nature. To address it, a deep learning framework based on auto-encoder, termed DeepAE, is proposed to elucidate high-dimensional transcriptomic profiling data in an encode-decode manner. Comparative experiments were conducted on nine transcriptomic profiling datasets to compare DeepAE with four benchmark methods. The results demonstrate that the proposed DeepAE outperforms the benchmark methods with robust performance on uncovering the key dimensions of single-cell RNA-seq data. In addition, we also investigate the performance of DeepAE in other contexts and platforms such as mass cytometry and metabolic profiling in a comprehensive manner. Gene ontology enrichment and pathology analysis are conducted to reveal the mechanisms behind the robust performance of DeepAE by uncovering its key dimensions., (© The Author(s) 2020. Published by Oxford University Press on behalf of Nucleic Acids Research.)

Motivation: The prediction of off-target mutations in CRISPR-Cas9 is a hot topic due to its relevance to gene editing research. Existing prediction methods have been developed; however, most of them just calculated scores based on mismatches to the guide sequence in CRISPR-Cas9. Therefore, the existing prediction methods are unable to scale and improve their performance with the rapid expansion of experimental data in CRISPR-Cas9. Moreover, the existing methods still cannot satisfy enough precision in off-target predictions for gene editing at the clinical level., Results: To address it, we design and implement two algorithms using deep neural networks to predict off-target mutations in CRISPR-Cas9 gene editing (i.e. deep convolutional neural network and deep feedforward neural network). The models were trained and tested on the recently released off-target dataset, CRISPOR dataset, for performance benchmark. Another off-target dataset identified by GUIDE-seq was adopted for additional evaluation. We demonstrate that convolutional neural network achieves the best performance on CRISPOR dataset, yielding an average classification area under the ROC curve (AUC) of 97.2% under stratified 5-fold cross-validation. Interestingly, the deep feedforward neural network can also be competitive at the average AUC of 97.0% under the same setting. We compare the two deep neural network models with the state-of-the-art off-target prediction methods (i.e. CFD, MIT, CROP-IT, and CCTop) and three traditional machine learning models (i.e. random forest, gradient boosting trees, and logistic regression) on both datasets in terms of AUC values, demonstrating the competitive edges of the proposed algorithms. Additional analyses are conducted to investigate the underlying reasons from different perspectives., Availability and Implementation: The example code are available at https://github.com/MichaelLinn/off_target_prediction. The related datasets are available at https://github.com/MichaelLinn/off_target_prediction/tree/master/data.

Bioactive peptide therapeutics has been a long-standing research topic. Notably, the antimicrobial peptides (AMPs) have been extensively studied for its therapeutic potential. Meanwhile, the demand for annotating other therapeutic peptides, such as antiviral peptides (AVPs) and anticancer peptides (ACPs), also witnessed an increase in recent years. However, we conceive that the structure of peptide chains and the intrinsic information between the amino acids is not fully investigated among the existing protocols. Therefore, we develop a new graph deep learning model, namely TP-LMMSG, which offers lightweight and easy-to-deploy advantages while improving the annotation performance in a generalizable manner. The results indicate that our model can accurately predict the properties of different peptides. The model surpasses the other state-of-the-art models on AMP, AVP and ACP prediction across multiple experimental validated datasets. Moreover, TP-LMMSG also addresses the challenges of time-consuming pre-processing in graph neural network frameworks. With its flexibility in integrating heterogeneous peptide features, our model can provide substantial impacts on the screening and discovery of therapeutic peptides. The source code is available at https://github.com/NanjunChen37/TP_LMMSG. [ABSTRACT FROM AUTHOR]

Protein acetylation is one of the extensively studied post-translational modifications (PTMs) due to its significant roles across a myriad of biological processes. Although many computational tools for acetylation site identification have been developed, there is a lack of benchmark dataset and bespoke predictors for non-histone acetylation site prediction. To address these problems, we have contributed to both dataset creation and predictor benchmark in this study. First, we construct a non-histone acetylation site benchmark dataset, namely NHAC, which includes 11 subsets according to the sequence length ranging from 11 to 61 amino acids. There are totally 886 positive samples and 4707 negative samples for each sequence length. Secondly, we propose TransPTM, a transformer-based neural network model for non-histone acetylation site predication. During the data representation phase, per-residue contextualized embeddings are extracted using ProtT5 (an existing pre-trained protein language model). This is followed by the implementation of a graph neural network framework, which consists of three TransformerConv layers for feature extraction and a multilayer perceptron module for classification. The benchmark results reflect that TransPTM has the competitive performance for non-histone acetylation site prediction over three state-of-the-art tools. It improves our comprehension on the PTM mechanism and provides a theoretical basis for developing drug targets for diseases. Moreover, the created PTM datasets fills the gap in non-histone acetylation site datasets and is beneficial to the related communities. The related source code and data utilized by TransPTM are accessible at https://www.github.com/TransPTM/TransPTM. [ABSTRACT FROM AUTHOR]

Single‐cell Hi‐C (scHi‐C) has made it possible to analyze chromatin organization at the single‐cell level. However, scHi‐C experiments generate inherently sparse data, which poses a challenge for loop calling methods. The existing approach performs significance tests across the imputed dense contact maps, leading to substantial computational overhead and loss of information at the single‐cell level. To overcome this limitation, a lightweight framework called scGSLoop is proposed, which sets a new paradigm for scHi‐C loop calling by adapting the training and inferencing strategies of graph‐based deep learning to leverage the sequence features and 1D positional information of genomic loci. With this framework, sparsity is no longer a challenge, but rather an advantage that the model leverages to achieve unprecedented computational efficiency. Compared to existing methods, scGSLoop makes more accurate predictions and is able to identify more loops that have the potential to play regulatory roles in genome functioning. Moreover, scGSLoop preserves single‐cell information by identifying a distinct group of loops for each individual cell, which not only enables an understanding of the variability of chromatin looping states between cells, but also allows scGSLoop to be extended for the investigation of multi‐connected hubs and their underlying mechanisms. [ABSTRACT FROM AUTHOR]

The rapid growth of omics-based data has revolutionized biomedical research and precision medicine, allowing machine learning models to be developed for cutting-edge performance. However, despite the wealth of high-throughput data available, the performance of these models is hindered by the lack of sufficient training data, particularly in clinical research (in vivo experiments). As a result, translating this knowledge into clinical practice, such as predicting drug responses, remains a challenging task. Transfer learning is a promising tool that bridges the gap between data domains by transferring knowledge from the source to the target domain. Researchers have proposed transfer learning to predict clinical outcomes by leveraging pre-clinical data (mouse, zebrafish), highlighting its vast potential. In this work, we present a comprehensive literature review of deep transfer learning methods for health informatics and clinical decision-making, focusing on high-throughput molecular data. Previous reviews mostly covered image-based transfer learning works, while we present a more detailed analysis of transfer learning papers. Furthermore, we evaluated original studies based on different evaluation settings across cross-validations, data splits and model architectures. The result shows that those transfer learning methods have great potential; high-throughput sequencing data and state-of-the-art deep learning models lead to significant insights and conclusions. Additionally, we explored various datasets in transfer learning papers with statistics and visualization. [ABSTRACT FROM AUTHOR]

Motivation Chromothripsis, associated with poor clinical outcomes, is prognostically vital in multiple myeloma. The catastrophic event is reported to be detectable prior to the progression of multiple myeloma. As a result, chromothripsis detection can contribute to risk estimation and early treatment guidelines for multiple myeloma patients. However, manual diagnosis remains the gold standard approach to detect chromothripsis events with the whole-genome sequencing technology to retrieve both copy number variation (CNV) and structural variation data. Meanwhile, CNV data are much easier to obtain than structural variation data. Hence, in order to reduce the reliance on human experts' efforts and structural variation data extraction, it is necessary to establish a reliable and accurate chromothripsis detection method based on CNV data. Results To address those issues, we propose a method to detect chromothripsis solely based on CNV data. With the help of structure learning, the intrinsic relationship-directed acyclic graph of CNV features is inferred to derive a CNV embedding graph (i.e. CNV-DAG). Subsequently, a neural network based on Graph Transformer, local feature extraction, and non-linear feature interaction, is proposed with the embedding graph as the input to distinguish whether the chromothripsis event occurs. Ablation experiments, clustering, and feature importance analysis are also conducted to enable the proposed model to be explained by capturing mechanistic insights. Availability and implementation The source code and data are freely available at https://github.com/luvyfdawnYu/CNV_chromothripsis. [ABSTRACT FROM AUTHOR]

Enhancers are short non-coding DNA sequences outside of the target promoter regions that can be bound by specific proteins to increase a gene's transcriptional activity, which has a crucial role in the spatiotemporal and quantitative regulation of gene expression. However, enhancers do not have a specific sequence motifs or structures, and their scattered distribution in the genome makes the identification of enhancers from human cell lines particularly challenging. Here we present a novel, stacked multivariate fusion framework called SMFM, which enables a comprehensive identification and analysis of enhancers from regulatory DNA sequences as well as their interpretation. Specifically, to characterize the hierarchical relationships of enhancer sequences, multi-source biological information and dynamic semantic information are fused to represent regulatory DNA enhancer sequences. Then, we implement a deep learning–based sequence network to learn the feature representation of the enhancer sequences comprehensively and to extract the implicit relationships in the dynamic semantic information. Ultimately, an ensemble machine learning classifier is trained based on the refined multi-source features and dynamic implicit relations obtained from the deep learning-based sequence network. Benchmarking experiments demonstrated that SMFM significantly outperforms other existing methods using several evaluation metrics. In addition, an independent test set was used to validate the generalization performance of SMFM by comparing it to other state-of-the-art enhancer identification methods. Moreover, we performed motif analysis based on the contribution scores of different bases of enhancer sequences to the final identification results. Besides, we conducted interpretability analysis of the identified enhancer sequences based on attention weights of EnhancerBERT, a fine-tuned BERT model that provides new insights into exploring the gene semantic information likely to underlie the discovered enhancers in an interpretable manner. Finally, in a human placenta study with 4,562 active distal gene regulatory enhancers, SMFM successfully exposed tissue-related placental development and the differential mechanism, demonstrating the generalizability and stability of our proposed framework. Author summary: Numerous evidence suggest that genes regulated by enhancers located in non-coding DNA regions are involved in a myriad of biological activities. To fully understand the regulatory role and mechanisms of enhancers on genes, the localization and identification of enhancers is essential. Several experimental biological methods are capable of localizing enhancers, however, these methods are resource intensive. To address this limitation, we developed a stacked multivariate fusion framework, called SMFM to identify and analyze enhancers with high accuracy and efficiency based on enhancer-specific dynamic semantic information and multi-source biological properties. The performance of the model is verified by experiments comparing different feature algorithms and classification algorithms. The superiority of our method is demonstrated by comparing it with several state-of-the-art algorithms. In addition, several analytical experiments demonstrate that SMFM is capable of recognizing enhancers in different tissues and detecting motifs in enhancers. To the best of our knowledge, this is the first computational approach that uses enhancer-specific dynamic semantic information to identify enhancers from regulatory DNA sequences and interpret them. It is expected that the SMFM model will effectively target enhancers and provide valid candidates for further biological experiments. [ABSTRACT FROM AUTHOR]

Healthcare disparities in multiethnic medical data is a major challenge; the main reason lies in the unequal data distribution of ethnic groups among data cohorts. Biomedical data collected from different cancer genome research projects may consist of mainly one ethnic group, such as people with European ancestry. In contrast, the data distribution of other ethnic races such as African, Asian, Hispanic, and Native Americans can be less visible than the counterpart. Data inequality in the biomedical field is an important research problem, resulting in the diverse performance of machine learning models while creating healthcare disparities. Previous researches have reduced the healthcare disparities only using limited data distributions. In our study, we work on fine-tuning of deep learning and transfer learning models with different multiethnic data distributions for the prognosis of 33 cancer types. In previous studies, to reduce the healthcare disparities, only a single ethnic cohort was used as the target domain with one major source domain. In contrast, we focused on multiple ethnic cohorts as the target domain in transfer learning using the TCGA and MMRF CoMMpass study datasets. After performance comparison for experiments with new data distributions, our proposed model shows promising performance for transfer learning schemes compared to the baseline approach for old and new data distributation experiments. [ABSTRACT FROM AUTHOR]

Background: Cancer molecular subtyping plays a critical role in individualized patient treatment. In previous studies, high-throughput gene expression signature-based methods have been proposed to identify cancer subtypes. Unfortunately, the existing ones suffer from the curse of dimensionality, data sparsity, and computational deficiency. Methods: To address those problems, we propose a computational framework for colorectal cancer subtyping without any exploitation in model complexity and generality. A supervised learning framework based on deep learning (DeepCSD) is proposed to identify cancer subtypes. Specifically, based on the differentially expressed genes under cancer consensus molecular subtyping, we design a minimalist feed-forward neural network to capture the distinct molecular features in different cancer subtypes. To mitigate the overfitting phenomenon of deep learning as much as possible, L1 and L2 regularization and dropout layers are added. Results: For demonstrating the effectiveness of DeepCSD, we compared it with other methods including Random Forest (RF), Deep forest (gcForest), support vector machine (SVM), XGBoost, and DeepCC on eight independent colorectal cancer datasets. The results reflect that DeepCSD can achieve superior performance over other algorithms. In addition, gene ontology enrichment and pathology analysis are conducted to reveal novel insights into the cancer subtype identification and characterization mechanisms. Conclusions: DeepCSD considers all subtype-specific genes as input, which is pathologically necessary for its completeness. At the same time, DeepCSD shows remarkable robustness in handling cross-platform gene expression data, achieving similar performance on both training and test data without significant model overfitting or exploitation of model complexity. [ABSTRACT FROM AUTHOR]

Identifying genome-wide binding events between circular RNAs (circRNAs) and RNA-binding proteins (RBPs) can greatly facilitate our understanding of functional mechanisms within circRNAs. Thanks to the development of cross-linked immunoprecipitation sequencing technology, large amounts of genome-wide circRNA binding event data have accumulated, providing opportunities for designing high-performance computational models to discriminate RBP interaction sites and thus to interpret the biological significance of circRNAs. Unfortunately, there are still no computational models sufficiently flexible to accommodate circRNAs from different data scales and with various degrees of feature representation. Here, we present HCRNet, a novel end-to-end framework for identification of circRNA-RBP binding events. To capture the hierarchical relationships, the multi-source biological information is fused to represent circRNAs, including various natural language sequence features. Furthermore, a deep temporal convolutional network incorporating global expectation pooling was developed to exploit the latent nucleotide dependencies in an exhaustive manner. We benchmarked HCRNet on 37 circRNA datasets and 31 linear RNA datasets to demonstrate the effectiveness of our proposed method. To evaluate further the model's robustness, we performed HCRNet on a full-length dataset containing 740 circRNAs. Results indicate that HCRNet generally outperforms existing methods. In addition, motif analyses were conducted to exhibit the interpretability of HCRNet on circRNAs. All supporting source code and data can be downloaded from https://github.com/yangyn533/HCRNet and https://doi.org/10.6084/m9.figshare.16943722.v1. And the web server of HCRNet is publicly accessible at http://39.104.118.143:5001/. [ABSTRACT FROM AUTHOR]

Motivation The rapid growth in literature accumulates diverse and yet comprehensive biomedical knowledge hidden to be mined such as drug interactions. However, it is difficult to extract the heterogeneous knowledge to retrieve or even discover the latest and novel knowledge in an efficient manner. To address such a problem, we propose EGFI for extracting and consolidating drug interactions from large-scale medical literature text data. Specifically, EGFI consists of two parts: classification and generation. In the classification part, EGFI encompasses the language model BioBERT which has been comprehensively pretrained on biomedical corpus. In particular, we propose the multihead self-attention mechanism and packed BiGRU to fuse multiple semantic information for rigorous context modeling. In the generation part, EGFI utilizes another pretrained language model BioGPT-2 where the generation sentences are selected based on filtering rules. Results We evaluated the classification part on 'DDIs 2013' dataset and 'DTIs' dataset, achieving the F1 scores of 0.842 and 0.720 respectively. Moreover, we applied the classification part to distinguish high-quality generated sentences and verified with the existing growth truth to confirm the filtered sentences. The generated sentences that are not recorded in DrugBank and DDIs 2013 dataset demonstrated the potential of EGFI to identify novel drug relationships. Availability Source code are publicly available at https://github.com/Layne-Huang/EGFI. [ABSTRACT FROM AUTHOR]

With the rapid increase in sequencing data, human host status inference (e.g. healthy or sick) from microbiome data has become an important issue. Existing studies are mostly based on single-point microbiome composition, while it is rare that the host status is predicted from longitudinal microbiome data. However, single-point-based methods cannot capture the dynamic patterns between the temporal changes and host status. Therefore, it remains challenging to build good predictive models as well as scaling to different microbiome contexts. On the other hand, existing methods are mainly targeted for disease prediction and seldom investigate other host statuses. To fill the gap, we propose a comprehensive deep learning-based framework that utilizes longitudinal microbiome data as input to infer the human host status. Specifically, the framework is composed of specific data preparation strategies and a recurrent neural network tailored for longitudinal microbiome data. In experiments, we evaluated the proposed method on both semi-synthetic and real datasets based on different sequencing technologies and metagenomic contexts. The results indicate that our method achieves robust performance compared to other baseline and state-of-the-art classifiers and provides a significant reduction in prediction time. [ABSTRACT FROM AUTHOR]

Circular RNAs (circRNAs) are widely expressed in eukaryotes. The genome-wide interactions between circRNAs and RNA-binding proteins (RBPs) can be probed from cross-linking immunoprecipitation with sequencing data. Therefore, computational methods have been developed for identifying RBP binding sites on circRNAs. Unfortunately, those computational methods often suffer from the low discriminative power of feature representations, numerical instability and poor scalability. To address those limitations, we propose a novel computational method called iCircRBP-DHN using deep hierarchical network for discriminating circRNA-RBP binding sites. The network architecture can be regarded as a deep multi-scale residual network followed by bidirectional gated recurrent units (BiGRUs) with the self-attention mechanism, which can simultaneously extract local and global contextual information. Meanwhile, we propose novel encoding schemes by integrating CircRNA2Vec and the K -tuple nucleotide frequency pattern to represent different degrees of nucleotide dependencies. To validate the effectiveness of our proposed iCircRBP-DHN, we compared its performance with other computational methods on 37 circRNAs datasets and 31 linear RNAs datasets, respectively. The experimental results reveal that iCircRBP-DHN can achieve superior performance over those state-of-the-art algorithms. Moreover, we perform motif analysis on circRNAs bound by those different RBPs, demonstrating that our proposed CircRNA2Vec encoding scheme can be promising. The iCircRBP-DHN method is made available at https://github.com/houzl3416/iCircRBP-DHN. [ABSTRACT FROM AUTHOR]

The off‐target effects induced by guide RNAs in the CRISPR/Cas9 gene‐editing system have raised substantial concerns in recent years. Many in silico predictive models have been developed for predicting the off‐target activities; however, few are capable of predicting the off‐target activities with insertions or deletions between guide RNA and target DNA sequence pair. In order to fill this gap, a recurrent convolutional network named CRISPR‐Net is developed for scoring the gRNA‐target pairs with mismatches and indels; and a machine‐learning based model named CRISPR‐Net‐Aggregate is also developed for aggregating the scores as the consensus off‐target score for each potential guide RNA. It is demonstrated that CRISPR‐Net achieves competitive performance on CIRCLE‐Seq and GUIDE‐seq datasets with indels and mismatches, outperforming the state‐of‐the‐art off‐target prediction methods on two independent mismatch‐only datasets. The CRISPR‐Net‐Aggregate also surpasses a competing method on the aggregation task. Moreover, a two‐stage sensitivity analysis is introduced to visualize the CRISPR‐Net prediction on the gRNA‐target pair of interest, demonstrating how implicit knowledge encoded in CRISPR‐Net contributes to the accurate off‐target activity quantification. Finally, the source code is made available at the Code Ocean repository (https://codeocean.com/capsule/9553651/tree/v1). [ABSTRACT FROM AUTHOR]

The analysis of animal movement from telemetry data provides insights into how and why animals move. While traditional approaches to such analysis mostly focus on predicting animal states during movement, we describe an approach that allows us to identify representative movement patterns of different animal groups. To do this, we propose a carefully designed recurrent neural network and combine it with telemetry data for automatic feature extraction and identification of non-predefined representative patterns. In the experiment, we consider a particular marine predator species, the southern elephant seal, as an example. With our approach, we identify that the male seals in our data set share similar movement patterns when they are close to land. We identify this pattern recurring in a number of distant locations, consistent with alternative approaches from previous research. [ABSTRACT FROM AUTHOR]

Cyber security has received increasing attention, as people use more Internet applications in their lives and worry about the security of their personal data on the Internet. Intrusion Detection Systems (IDSs) are critical security tools that can detect and respond to intrusions. In recent years, Deep Learning (DL) techniques have gained popularity in IDS design due to their promising performance in terms of detection accuracy. However, the design of DL architectures usually requires professional knowledge and significantly impacts the performance of the DL model. Furthermore, the existence of a small ratio of abnormal traffic in vast network traffic leads to a serious imbalanced data problem, which negatively affects the performance of the DL model in detecting minority attack classes. To alleviate these problems, this paper proposes a multi-objective evolutionary DL model (called EvoBMF) to detect network intrusion behaviors. The model incorporates bidirectional Long–short Term Memory (BiLSTM) for preliminary feature extraction, Multi-Head Attention (MHA) for further capturing features and global information of the network traffic, and Full-Connected Layer (FCL) module to perform final classification. To deal with the challenge of manually tuning the parameters of the DL model when tackling different tasks, the parameters of the EvoBMF model are first encoded as the chromosome of the Multi-objective Evolutionary Algorithm (MOEA), which aims to optimize the two conflicting objectives (complexity and classification ability) of the model. A state-of-the-art MOEA (MOEA/D-DRA) is then used to optimize the above two objectives, aiming to obtain the optimal architecture for EvoBMF, which can be easily deployed in cloud computing scenarios to detect and respond to network intrusions. Additionally, to alleviate the severe imbalance in routine network traffic, the synthetic minority over-sampling technique is introduced to generate representative samples of minority classes to improve the overall performance of the model. At last, the experimental results conducted on two popular datasets (UNSW-NB15 and CIC-IDS 2018) have demonstrated that the proposed EvoBMF model can provide superior performance for intrusion detection when compared to some state-of-the-art IDSs. • EvoBMF: a multi-objective evolutionary DL model for IDS. • EvoBMF uses BiLSTM, MHA, and FCL for feature extraction and classification. • NAS technique automates parameter adjustment of EvoBMF. • SMOTE technique improves recognition rate of minority attack classes. [ABSTRACT FROM AUTHOR]

Our world is moving fast towards the era of the Internet of Things (IoT), which connects all kinds of devices to digital services and brings significant convenience to our lives. With the rapid increase in the number of devices connected to the IoT, there may exist more network vulnerabilities, resulting in more network attacks. Under this dynamic IoT environment, an effective intrusion detection system (IDS) is urgently needed to detect attacks with low-latency and high accuracy. A number of promising IDSs have been proposed based on deep learning (DL) techniques, but they need to do parameter tuning under different environments, which is very time-consuming. To alleviate this problem, this paper proposes a multi-objective evolutionary convolutional neural network for intrusion detection system, called MECNN, which is run on the fog nodes of Fog computing on IoT. In this approach, convolutional neural network (CNN) is used as the classifier to detect intrusions and the multi-objective evolutionary algorithm based on decomposition (MOEA/D) algorithm is modified to evolve the CNN model, which greatly simplifies the parameter tuning process of DL. To be specific, a novel encoding scheme is first proposed to transform the topological architecture of CNN into a chromosome of MOEA/D and then the two conflicting objectives, i.e., detection performance and model complexity of the CNN model, are simultaneously optimized by MOEA/D, which can obtain a number of IDSs with various detection performance and model complexities. Then, the most suitable MECNN model can be deployed in different fog nodes of Fog computing, providing low-latency and high-accuracy intrusion detection for IoT. Finally, the experimental studies are conducted on two popular datasets (AWID and CIC-IDS2107), which have validated that our MECNN model can improve detection performance and robustness to better protect the IoT when compared to other state-of-the-art IDSs. [ABSTRACT FROM AUTHOR]