Deep Learning for Genomics: From Early Neural Nets to Modern Large Language Models

The data explosion driven by advancements in genomic research, such as high-throughput sequencing techniques, is constantly challenging conventional methods used in genomics. In parallel with the urgent demand for robust algorithms, deep learning has succeeded in various fields such as vision, speech, and text processing. Yet genomics entails unique challenges to deep learning, since we expect a superhuman intelligence that explores beyond our knowledge to interpret the genome from deep learning. A powerful deep learning model should rely on the insightful utilization of task-specific knowledge. In this paper, we briefly discuss the strengths of different deep learning models from a genomic perspective so as to fit each particular task with proper deep learning-based architecture, and we remark on practical considerations of developing deep learning architectures for genomics. We also provide a concise review of deep learning applications in various aspects of genomic research and point out current challenges and potential research directions for future genomics applications. We believe the collaborative use of ever-growing diverse data and the fast iteration of deep learning models will continue to contribute to the future of genomics.

Identifying Tissue- and Cohort-Specific RNA Regulatory Modules in Cancer Cells Using Multitask Learning

Simple Summary

Understanding the underlying biological mechanisms of primary tumors is crucial for predicting how tumors respond to therapies and exploring accurate treatment strategies. miRNA–mRNA interactions have a major effect on many biological processes that are important in the formation and progression of cancer. In this study, we introduced a computational pipeline to extract tissue- and cohort-specific miRNA–mRNA regulatory modules of multiple cancer types from the same origin using miRNA and mRNA expression profiles of primary tumors. Our model identified regulatory modules of underlying cancer types (i.e., cohort-specific) and shared regulatory modules between cohorts (i.e., tissue-specific).

Abstract

MicroRNA (miRNA) alterations significantly impact the formation and progression of human cancers. miRNAs interact with messenger RNAs (mRNAs) to facilitate degradation or translational repression. Thus, identifying miRNA–mRNA regulatory modules in cohorts of primary tumor tissues are fundamental for understanding the biology of tumor heterogeneity and precise diagnosis and treatment. We established a multitask learning sparse regularized factor regression (MSRFR) method to determine key tissue- and cohort-specific miRNA–mRNA regulatory modules from expression profiles of tumors. MSRFR simultaneously models the sparse relationship between miRNAs and mRNAs and extracts tissue- and cohort-specific miRNA–mRNA regulatory modules separately. We tested the model’s ability to determine cohort-specific regulatory modules of multiple cancer cohorts from the same tissue and their underlying tissue-specific regulatory modules by extracting similarities between cancer cohorts (i.e., blood, kidney, and lung). We also detected tissue-specific and cohort-specific signatures in the corresponding regulatory modules by comparing our findings from various other tissues. We show that MSRFR effectively determines cancer-related miRNAs in cohort-specific regulatory modules, distinguishes tissue- and cohort-specific regulatory modules from each other, and extracts tissue-specific information from different cohorts of disease-related tissue. Our findings indicate that the MSRFR model can support current efforts in precision medicine to define tumor-specific miRNA–mRNA signatures.

RBP-TSTL is a two-stage transfer learning framework for genome-scale prediction of RNA-binding proteins

RNA binding proteins (RBPs) are critical for the post-transcriptional control of RNAs and play vital roles in a myriad of biological processes, such as RNA localization and gene regulation. Therefore, computational methods that are capable of accurately identifying RBPs are highly desirable and have important implications for biomedical and biotechnological applications. Here, we propose a two-stage deep transfer learning-based framework, termed RBP-TSTL, for accurate prediction of RBPs. In the first stage, the knowledge from the self-supervised pre-trained model was extracted as feature embeddings and used to represent the protein sequences, while in the second stage, a customized deep learning model was initialized based on an annotated pre-training RBPs dataset before being fine-tuned on each corresponding target species dataset. This two-stage transfer learning framework can enable the RBP-TSTL model to be effectively trained to learn and improve the prediction performance. Extensive performance benchmarking of the RBP-TSTL models trained using the features generated by the self-supervised pre-trained model and other models trained using hand-crafting encoding features demonstrated the effectiveness of the proposed two-stage knowledge transfer strategy based on the self-supervised pre-trained models. Using the best-performing RBP-TSTL models, we further conducted genome-scale RBP predictions for Homo sapiens, Arabidopsis thaliana, Escherichia coli, and Salmonella and established a computational compendium containing all the predicted putative RBPs candidates. We anticipate that the proposed RBP-TSTL approach will be explored as a useful tool for the characterization of RNA-binding proteins and exploration of their sequence–structure–function relationships.

A Machine Learning Strategy Based on Kittler’s Taxonomy to Detect Anomalies and Recognize Contexts Applied to Monitor Water Bodies in Environments

Environmental monitoring, such as analyses of water bodies to detect anomalies, is recognized worldwide as a task necessary to reduce the impacts arising from pollution. However, the large number of data available to be analyzed in different contexts, such as in an image time series acquired by satellites, still pose challenges for the detection of anomalies, even when using computers. This study describes a machine learning strategy based on Kittler’s taxonomy to detect anomalies related to water pollution in an image time series. We propose this strategy to monitor environments, detecting unexpected conditions that may occur (i.e., detecting outliers), and identifying those outliers in accordance with Kittler’s taxonomy (i.e., detecting anomalies). According to our strategy, contextual and non-contextual image classifications were semi-automatically compared to find any divergence that indicates the presence of one type of anomaly defined by the taxonomy. In our strategy, models built to classify a single image were used to classify an image time series due to domain adaptation. The results 99.07%, 99.99%, 99.07%, and 99.53% were achieved by our strategy, respectively, for accuracy, precision, recall, and F-measure. These results suggest that our strategy allows computers to recognize contexts and enhances their capabilities to solve contextualized problems. Therefore, our strategy can be used to guide computational systems to make different decisions to solve a problem in response to each context. The proposed strategy is relevant for improving machine learning, as its use allows computers to have a more organized learning process. Our strategy is presented with respect to its applicability to help monitor environmental disasters. A minor limitation was found in the results caused by the use of domain adaptation. This type of limitation is fairly common when using domain adaptation, and therefore has no significance. Even so, future work should investigate other techniques for transfer learning.

Evolutionary Multitasking for Multiobjective Optimization With Subspace Alignment and Adaptive Differential Evolution

In contrast to the traditional single-tasking evolutionary algorithms, evolutionary multitasking (EMT) travels in the search space of multiple optimization tasks simultaneously. Through sharing knowledge across the tasks, EMT is able to enhance solving the optimization tasks. However, if knowledge transfer is not properly carried out, the performance of EMT might become unsatisfactory. To address this issue and improve the quality of knowledge transfer among the tasks, a novel multiobjective EMT algorithm based on subspace alignment and self-adaptive differential evolution (DE), namely, MOMFEA-SADE, is proposed in this article. Particularly, a mapping matrix obtained by subspace learning is used to transform the search space of the population and reduce the probability of negative knowledge transfer between tasks. In addition, DE characterized by a self-adaptive trial vector generation strategy is introduced to generate promising solutions based on previous experiences. The experimental results on multiobjective multi/many-tasking optimization test suites show that MOMFEA-SADE is superior or comparable to other state-of-the-art EMT algorithms. MOMFEA-SADE also won the Competition on Evolutionary Multitask Optimization (the multitask multiobjective optimization track) within IEEE 2019 Congress on Evolutionary Computation.

xDeep-AcPEP: Deep Learning Method for Anticancer Peptide Activity Prediction Based on Convolutional Neural Network and Multitask Learning

Cancer is one of the leading causes of death worldwide. Conventional cancer treatment relies on radiotherapy and chemotherapy, but both methods bring severe side effects to patients, as these therapies not only attack cancer cells but also damage normal cells. Anticancer peptides (ACPs) are a promising alternative as therapeutic agents that are efficient and selective against tumor cells. Here, we propose a deep learning method based on convolutional neural networks to predict biological activity (EC50, LC50, IC50, and LD50) against six tumor cells, including breast, colon, cervix, lung, skin, and prostate. We show that models derived with multitask learning achieve better performance than conventional single-task models. In repeated 5-fold cross validation using the CancerPPD data set, the best models with the applicability domain defined obtain an average mean squared error of 0.1758, Pearson's correlation coefficient of 0.8086, and Kendall's correlation coefficient of 0.6156. As a step toward model interpretability, we infer the contribution of each residue in the sequence to the predicted activity by means of feature importance weights derived from the convolutional layers of the model. The present method, referred to as xDeep-AcPEP, will help to identify effective ACPs in rational peptide design for therapeutic purposes. The data, script files for reproducing the experiments, and the final prediction models can be downloaded from http://github.com/chen709847237/xDeep-AcPEP. The web server to directly access this prediction method is at https://app.cbbio.online/acpep/home.

A Review of Domain Adaptation without Target Labels

Domain adaptation has become a prominent problem setting in machine learning and related fields. This review asks the question: How can a classifier learn from a source domain and generalize to a target domain? We present a categorization of approaches, divided into, what we refer to as, sample-based, feature-based, and inference-based methods. Sample-based methods focus on weighting individual observations during training based on their importance to the target domain. Feature-based methods revolve around on mapping, projecting, and representing features such that a source classifier performs well on the target domain and inference-based methods incorporate adaptation into the parameter estimation procedure, for instance through constraints on the optimization procedure. Additionally, we review a number of conditions that allow for formulating bounds on the cross-domain generalization error. Our categorization highlights recurring ideas and raises questions important to further research.

A multitask multiple kernel learning formulation for discriminating early- and late-stage cancers

Motivation

Genomic information is increasingly being used in diagnosis, prognosis and treatment of cancer. The severity of the disease is usually measured by the tumor stage. Therefore, identifying pathways playing an important role in progression of the disease stage is of great interest. Given that there are similarities in the underlying mechanisms of different cancers, in addition to the considerable correlation in the genomic data, there is a need for machine learning methods that can take these aspects of genomic data into account. Furthermore, using machine learning for studying multiple cancer cohorts together with a collection of molecular pathways creates an opportunity for knowledge extraction.

Results

We studied the problem of discriminating early- and late-stage tumors of several cancers using genomic information while enforcing interpretability on the solutions. To this end, we developed a multitask multiple kernel learning (MTMKL) method with a co-clustering step based on a cutting-plane algorithm to identify the relationships between the input tasks and kernels. We tested our algorithm on 15 cancer cohorts and observed that, in most cases, MTMKL outperforms other algorithms (including random forests, support vector machine and single-task multiple kernel learning) in terms of predictive power. Using the aggregate results from multiple replications, we also derived similarity matrices between cancer cohorts, which are, in many cases, in agreement with available relationships reported in the relevant literature.

Availability and implementation

Our implementations of support vector machine and multiple kernel learning algorithms in R are available at https://github.com/arezourahimi/mtgsbc together with the scripts that replicate the reported experiments.

Supplementary information

Supplementary data are available at Bioinformatics online.

Multi-modal neuroimaging feature selection with consistent metric constraint for diagnosis of Alzheimer's disease

The accurate diagnosis of Alzheimer's disease (AD) and its early stage, e.g., mild cognitive impairment (MCI), is essential for timely treatment or possible intervention to slow down AD progression. Recent studies have demonstrated that multiple neuroimaging and biological measures contain complementary information for diagnosis and prognosis. Therefore, information fusion strategies with multi-modal neuroimaging data, such as voxel-based measures extracted from structural MRI (VBM-MRI) and fluorodeoxyglucose positron emission tomography (FDG-PET), have shown their effectiveness for AD diagnosis. However, most existing methods are proposed to simply integrate the multi-modal data, but do not make full use of structure information across the different modalities. In this paper, we propose a novel multi-modal neuroimaging feature selection method with consistent metric constraint (MFCC) for AD analysis. First, the similarity is calculated for each modality (i.e. VBM-MRI or FDG-PET) individually by random forest strategy, which can extract pairwise similarity measures for multiple modalities. Then the group sparsity regularization term and the sample similarity constraint regularization term are used to constrain the objective function to conduct feature selection from multiple modalities. Finally, the multi-kernel support vector machine (MK-SVM) is used to fuse the features selected from different models for final classification. The experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) show that the proposed method has better classification performance than the start-of-the-art multimodality-based methods. Specifically, we achieved higher accuracy and area under the curve (AUC) for AD versus normal controls (NC), MCI versus NC, and MCI converters (MCI-C) versus MCI non-converters (MCI-NC) on ADNI datasets. Therefore, the proposed model not only outperforms the traditional method in terms of AD/MCI classification, but also discovers the characteristics associated with the disease, demonstrating its promise for improving disease-related mechanistic understanding.

An Incongruence-Based Anomaly Detection Strategy for Analyzing Water Pollution in Images from Remote Sensing

The potential applications of computational tools, such as anomaly detection and incongruence, for analyzing data attract much attention from the scientific research community. However, there remains a need for more studies to determine how anomaly detection and incongruence applied to analyze data of static images from remote sensing will assist in detecting water pollution. In this study, an incongruence-based anomaly detection strategy for analyzing water pollution in images from remote sensing is presented. Our strategy semi-automatically detects occurrences of one type of anomaly based on the divergence between two image classifications (contextual and non-contextual). The results indicate that our strategy accurately analyzes the majority of images. Incongruence as a strategy for detecting anomalies in real-application (non-synthetic) data found in images from remote sensing is relevant for recognizing crude oil close to open water bodies or water pollution caused by the presence of brown mud in large rivers. It can also assist surveillance systems by detecting environmental disasters or performing mappings.

Targeting Virus-host Protein Interactions: Feature Extraction and Machine Learning Approaches

BACKGROUND

Targeting critical viral-host Protein-Protein Interactions (PPIs) has enormous application prospects for therapeutics. Using experimental methods to evaluate all possible virus-host PPIs is labor-intensive and time-consuming. Recent growth in computational identification of virus-host PPIs provides new opportunities for gaining biological insights, including applications in disease control. We provide an overview of recent computational approaches for studying virus-host PPI interactions.

METHODS

In this review, a variety of computational methods for virus-host PPIs prediction have been surveyed. These methods are categorized based on the features they utilize and different machine learning algorithms including classical and novel methods.

RESULTS

We describe the pivotal and representative features extracted from relevant sources of biological data, mainly include sequence signatures, known domain interactions, protein motifs and protein structure information. We focus on state-of-the-art machine learning algorithms that are used to build binary prediction models for the classification of virus-host protein pairs and discuss their abilities, weakness and future directions.

CONCLUSION

The findings of this review confirm the importance of computational methods for finding the potential protein-protein interactions between virus and host. Although there has been significant progress in the prediction of virus-host PPIs in recent years, there is a lot of room for improvement in virus-host PPI prediction.

Transfer String Kernel for Cross-Context DNA-Protein Binding Prediction

Through sequence-based classification, this paper tries to accurately predict the DNA binding sites of transcription factors (TFs) in an unannotated cellular context. Related methods in the literature fail to perform such predictions accurately, since they do not consider sample distribution shift of sequence segments from an annotated (source) context to an unannotated (target) context. We, therefore, propose a method called "Transfer String Kernel" (TSK) that achieves improved prediction of transcription factor binding site (TFBS) using knowledge transfer via cross-context sample adaptation. TSK maps sequence segments to a high-dimensional feature space using a discriminative mismatch string kernel framework. In this high-dimensional space, labeled examples of the source context are re-weighted so that the revised sample distribution matches the target context more closely. We have experimentally verified TSK for TFBS identifications on 14 different TFs under a cross-organism setting. We find that TSK consistently outperforms the state-of-the-art TFBS tools, especially when working with TFs whose binding sequences are not conserved across contexts. We also demonstrate the generalizability of TSK by showing its cutting-edge performance on a different set of cross-context tasks for the MHC peptide binding predictions.

Review of computational methods for virus-host protein interaction prediction: a case study on novel Ebola-human interactions

Identification of potential virus–host interactions is useful and vital to control the highly infectious virus-caused diseases. This may contribute toward development of new drugs to treat the viral infections. Recently, database records of clinically and experimentally validated interactions between a small set of human proteins and Ebola virus (EBOV) have been published. Using the information of the known human interaction partners of EBOV, our main objective is to identify a set of proteins that may interact with EBOV proteins. Here, we first review the state-of-the-art, computational methods used for prediction of novel virus–host interactions for infectious diseases followed by a case study on EBOV–human interactions. The assessment result shows that the predicted human host proteins are highly similar with known human interaction partners of EBOV in the context of structure and semantics and are responsible for similar biochemical activities, pathways and host–pathogen relationships.

Probabilistic Low-Rank Multitask Learning

In this paper, we consider the problem of learning multiple related tasks simultaneously with the goal of improving the generalization performance of individual tasks. The key challenge is to effectively exploit the shared information across multiple tasks as well as preserve the discriminative information for each individual task. To address this, we propose a novel probabilistic model for multitask learning (MTL) that can automatically balance between low-rank and sparsity constraints. The former assumes a low-rank structure of the underlying predictive hypothesis space to explicitly capture the relationship of different tasks and the latter learns the incoherent sparse patterns private to each task. We derive and perform inference via variational Bayesian methods. Experimental results on both regression and classification tasks on real-world applications demonstrate the effectiveness of the proposed method in dealing with the MTL problems.

A hybrid approach to automatic de-identification of psychiatric notes

De-identification, or identifying and removing protected health information (PHI) from clinical data, is a critical step in making clinical data available for clinical applications and research. This paper presents a natural language processing system for automatic de-identification of psychiatric notes, which was designed to participate in the 2016 CEGS N-GRID shared task Track 1. The system has a hybrid structure that combines machine leaning techniques and rule-based approaches. The rule-based components exploit the structure of the psychiatric notes as well as characteristic surface patterns of PHI mentions. The machine learning components utilize supervised learning with rich features. In addition, the system performance was boosted with integration of additional data to the training set through domain adaptation. The hybrid system showed overall micro-averaged F-score 90.74 on the test set, second-best among all the participants of the CEGS N-GRID task.

Ranking Medical Terms to Support Expansion of Lay Language Resources for Patient Comprehension of Electronic Health Record Notes: Adapted Distant Supervision Approach

Background

Medical terms are a major obstacle for patients to comprehend their electronic health record (EHR) notes. Clinical natural language processing (NLP) systems that link EHR terms to lay terms or definitions allow patients to easily access helpful information when reading through their EHR notes, and have shown to improve patient EHR comprehension. However, high-quality lay language resources for EHR terms are very limited in the public domain. Because expanding and curating such a resource is a costly process, it is beneficial and even necessary to identify terms important for patient EHR comprehension first.

Objective

We aimed to develop an NLP system, called adapted distant supervision (ADS), to rank candidate terms mined from EHR corpora. We will give EHR terms ranked as high by ADS a higher priority for lay language annotation—that is, creating lay definitions for these terms.

Methods

Adapted distant supervision uses distant supervision from consumer health vocabulary and transfer learning to adapt itself to solve the problem of ranking EHR terms in the target domain. We investigated 2 state-of-the-art transfer learning algorithms (ie, feature space augmentation and supervised distant supervision) and designed 5 types of learning features, including distributed word representations learned from large EHR data for ADS. For evaluating ADS, we asked domain experts to annotate 6038 candidate terms as important or nonimportant for EHR comprehension. We then randomly divided these data into the target-domain training data (1000 examples) and the evaluation data (5038 examples). We compared ADS with 2 strong baselines, including standard supervised learning, on the evaluation data.

Results

The ADS system using feature space augmentation achieved the best average precision, 0.850, on the evaluation set when using 1000 target-domain training examples. The ADS system using supervised distant supervision achieved the best average precision, 0.819, on the evaluation set when using only 100 target-domain training examples. The 2 ADS systems both performed significantly better than the baseline systems (P<.001 for all measures and all conditions). Using a rich set of learning features contributed to ADS’s performance substantially.

Conclusions

ADS can effectively rank terms mined from EHRs. Transfer learning improved ADS’s performance even with a small number of target-domain training examples. EHR terms prioritized by ADS were used to expand a lay language resource that supports patient EHR comprehension. The top 10,000 EHR terms ranked by ADS are available upon request.

VB-MK-LMF: fusion of drugs, targets and interactions using variational Bayesian multiple kernel logistic matrix factorization

BACKGROUND

Computational fusion approaches to drug-target interaction (DTI) prediction, capable of utilizing multiple sources of background knowledge, were reported to achieve superior predictive performance in multiple studies. Other studies showed that specificities of the DTI task, such as weighting the observations and focusing the side information are also vital for reaching top performance.

METHOD

We present Variational Bayesian Multiple Kernel Logistic Matrix Factorization (VB-MK-LMF), which unifies the advantages of (1) multiple kernel learning, (2) weighted observations, (3) graph Laplacian regularization, and (4) explicit modeling of probabilities of binary drug-target interactions.

RESULTS

VB-MK-LMF achieves significantly better predictive performance in standard benchmarks compared to state-of-the-art methods, which can be traced back to multiple factors. The systematic evaluation of the effect of multiple kernels confirm their benefits, but also highlights the limitations of linear kernel combinations, already recognized in other fields. The analysis of the effect of prior kernels using varying sample sizes sheds light on the balance of data and knowledge in DTI tasks and on the rate at which the effect of priors vanishes. This also shows the existence of "small sample size" regions where using side information offers significant gains. Alongside favorable predictive performance, a notable property of MF methods is that they provide a unified space for drugs and targets using latent representations. Compared to earlier studies, the dimensionality of this space proved to be surprisingly low, which makes the latent representations constructed by VB-ML-LMF especially well-suited for visual analytics. The probabilistic nature of the predictions allows the calculation of the expected values of hits in functionally relevant sets, which we demonstrate by predicting drug promiscuity. The variational Bayesian approximation is also implemented for general purpose graphics processing units yielding significantly improved computational time.

CONCLUSION

In standard benchmarks, VB-MK-LMF shows significantly improved predictive performance in a wide range of settings. Beyond these benchmarks, another contribution of our work is highlighting and providing estimates for further pharmaceutically relevant quantities, such as promiscuity, druggability and total number of interactions.

Knowledge-transfer learning for prediction of matrix metalloprotease substrate-cleavage sites

Matrix Metalloproteases (MMPs) are an important family of proteases that play crucial roles in key cellular and disease processes. Therefore, MMPs constitute important targets for drug design, development and delivery. Advanced proteomic technologies have identified type-specific target substrates; however, the complete repertoire of MMP substrates remains uncharacterized. Indeed, computational prediction of substrate-cleavage sites associated with MMPs is a challenging problem. This holds especially true when considering MMPs with few experimentally verified cleavage sites, such as for MMP-2, -3, -7, and -8. To fill this gap, we propose a new knowledge-transfer computational framework which effectively utilizes the hidden shared knowledge from some MMP types to enhance predictions of other, distinct target substrate-cleavage sites. Our computational framework uses support vector machines combined with transfer machine learning and feature selection. To demonstrate the value of the model, we extracted a variety of substrate sequence-derived features and compared the performance of our method using both 5-fold cross-validation and independent tests. The results show that our transfer-learning-based method provides a robust performance, which is at least comparable to traditional feature-selection methods for prediction of MMP-2, -3, -7, -8, -9 and -12 substrate-cleavage sites on independent tests. The results also demonstrate that our proposed computational framework provides a useful alternative for the characterization of sequence-level determinants of MMP-substrate specificity.

Domain adaptation for semantic role labeling of clinical text

OBJECTIVE

Semantic role labeling (SRL), which extracts a shallow semantic relation representation from different surface textual forms of free text sentences, is important for understanding natural language. Few studies in SRL have been conducted in the medical domain, primarily due to lack of annotated clinical SRL corpora, which are time-consuming and costly to build. The goal of this study is to investigate domain adaptation techniques for clinical SRL leveraging resources built from newswire and biomedical literature to improve performance and save annotation costs.

MATERIALS AND METHODS

Multisource Integrated Platform for Answering Clinical Questions (MiPACQ), a manually annotated SRL clinical corpus, was used as the target domain dataset. PropBank and NomBank from newswire and BioProp from biomedical literature were used as source domain datasets. Three state-of-the-art domain adaptation algorithms were employed: instance pruning, transfer self-training, and feature augmentation. The SRL performance using different domain adaptation algorithms was evaluated by using 10-fold cross-validation on the MiPACQ corpus. Learning curves for the different methods were generated to assess the effect of sample size.

RESULTS AND CONCLUSION

When all three source domain corpora were used, the feature augmentation algorithm achieved statistically significant higher F-measure (83.18%), compared to the baseline with MiPACQ dataset alone (F-measure, 81.53%), indicating that domain adaptation algorithms may improve SRL performance on clinical text. To achieve a comparable performance to the baseline method that used 90% of MiPACQ training samples, the feature augmentation algorithm required <50% of training samples in MiPACQ, demonstrating that annotation costs of clinical SRL can be reduced significantly by leveraging existing SRL resources from other domains.

Computational approaches for prediction of pathogen-host protein-protein interactions

Infectious diseases are still among the major and prevalent health problems, mostly because of the drug resistance of novel variants of pathogens. Molecular interactions between pathogens and their hosts are the key parts of the infection mechanisms. Novel antimicrobial therapeutics to fight drug resistance is only possible in case of a thorough understanding of pathogen-host interaction (PHI) systems. Existing databases, which contain experimentally verified PHI data, suffer from scarcity of reported interactions due to the technically challenging and time consuming process of experiments. These have motivated many researchers to address the problem by proposing computational approaches for analysis and prediction of PHIs. The computational methods primarily utilize sequence information, protein structure and known interactions. Classic machine learning techniques are used when there are sufficient known interactions to be used as training data. On the opposite case, transfer and multitask learning methods are preferred. Here, we present an overview of these computational approaches for predicting PHI systems, discussing their weakness and abilities, with future directions.

OncoSearch: cancer gene search engine with literature evidence

In order to identify genes that are involved in oncogenesis and to understand how such genes affect cancers, abnormal gene expressions in cancers are actively studied. For an efficient access to the results of such studies that are reported in biomedical literature, the relevant information is accumulated via text-mining tools and made available through the Web. However, current Web tools are not yet tailored enough to allow queries that specify how a cancer changes along with the change in gene expression level, which is an important piece of information to understand an involved gene's role in cancer progression or regression. OncoSearch is a Web-based engine that searches Medline abstracts for sentences that mention gene expression changes in cancers, with queries that specify (i) whether a gene expression level is up-regulated or down-regulated, (ii) whether a certain type of cancer progresses or regresses along with such gene expression change and (iii) the expected role of the gene in the cancer. OncoSearch is available through http://oncosearch.biopathway.org.

Multitask learning for host-pathogen protein interactions

Motivation: An important aspect of infectious disease research involves understanding the differences and commonalities in the infection mechanisms underlying various diseases. Systems biology-based approaches study infectious diseases by analyzing the interactions between the host species and the pathogen organisms. This work aims to combine the knowledge from experimental studies of host–pathogen interactions in several diseases to build stronger predictive models. Our approach is based on a formalism from machine learning called ‘multitask learning’, which considers the problem of building models across tasks that are related to each other. A ‘task’ in our scenario is the set of host–pathogen protein interactions involved in one disease. To integrate interactions from several tasks (i.e. diseases), our method exploits the similarity in the infection process across the diseases. In particular, we use the biological hypothesis that similar pathogens target the same critical biological processes in the host, in defining a common structure across the tasks.

Results: Our current work on host–pathogen protein interaction prediction focuses on human as the host, and four bacterial species as pathogens. The multitask learning technique we develop uses a task-based regularization approach. We find that the resulting optimization problem is a difference of convex (DC) functions. To optimize, we implement a Convex–Concave procedure-based algorithm. We compare our integrative approach to baseline methods that build models on a single host–pathogen protein interaction dataset. Our results show that our approach outperforms the baselines on the training data. We further analyze the protein interaction predictions generated by the models, and find some interesting insights.

Availability: The predictions and code are available at: http://www.cs.cmu.edu/∼mkshirsa/ismb2013_paper320.html

Contact:j.klein-seetharaman@warwick.ac.uk

Supplementary information:Supplementary data are available at Bioinformatics online.

Scoring functions for protein-protein interactions

The computational evaluation of protein-protein interactions will play an important role in organising the wealth of data being generated by high-throughput initiatives. Here we discuss future applications, report recent developments and identify areas requiring further investigation. Many functions have been developed to quantify the structural and energetic properties of interacting proteins, finding use in interrelated challenges revolving around the relationship between sequence, structure and binding free energy. These include loop modelling, side-chain refinement, docking, multimer assembly, affinity prediction, affinity change upon mutation, hotspots location and interface design. Information derived from models optimised for one of these challenges can be used to benefit the others, and can be unified within the theoretical frameworks of multi-task learning and Pareto-optimal multi-objective learning.