Multimodal digital assessment of depression with actigraphy and app in Hong Kong Chinese.

There is an emerging potential for digital assessment of depression. In this study, Chinese patients with major depressive disorder (MDD) and controls underwent a week of multimodal measurement including actigraphy and app-based measures (D-MOMO) to record rest-activity, facial expression, voice, and mood states. Seven machine-learning models (Random Forest [RF], Logistic regression [LR], Support vector machine [SVM], K-Nearest Neighbors [KNN], Decision tree [DT], Naive Bayes [NB], and Artificial Neural Networks [ANN]) with leave-one-out cross-validation were applied to detect lifetime diagnosis of MDD and non-remission status. Eighty MDD subjects and 76 age- and sex-matched controls completed the actigraphy, while 61 MDD subjects and 47 controls completed the app-based assessment. MDD subjects had lower mobile time (P = 0.006), later sleep midpoint (P = 0.047) and Acrophase (P = 0.024) than controls. For app measurement, MDD subjects had more frequent brow lowering (P = 0.023), less lip corner pulling (P = 0.007), higher pause variability (P = 0.046), more frequent self-reference (P = 0.024) and negative emotion words (P = 0.002), lower articulation rate (P < 0.001) and happiness level (P < 0.001) than controls. With the fusion of all digital modalities, the predictive performance (F1-score) of ANN for a lifetime diagnosis of MDD was 0.81 and 0.70 for non-remission status when combined with the HADS-D item score, respectively. Multimodal digital measurement is a feasible diagnostic tool for depression in Chinese. A combination of multimodal measurement and machine-learning approach has enhanced the performance of digital markers in phenotyping and diagnosis of MDD.

Detection of Suicidal Ideation in Clinical Interviews for Depression Using Natural Language Processing and Machine Learning: Cross-Sectional Study.

Background: Assessing patients' suicide risk is challenging, especially among those who deny suicidal ideation. Primary care providers have poor agreement in screening suicide risk. Patients' speech may provide more objective, language-based clues about their underlying suicidal ideation. Text analysis to detect suicide risk in depression is lacking in the literature. Objective: This study aimed to determine whether suicidal ideation can be detected via language features in clinical interviews for depression using natural language processing (NLP) and machine learning (ML). Methods: This cross-sectional study recruited 305 participants between October 2020 and May 2022 (mean age 53.0, SD 11.77 years; female: n=176, 57%), of which 197 had lifetime depression and 108 were healthy. This study was part of ongoing research on characterizing depression with a case-control design. In this study, 236 participants were nonsuicidal, while 56 and 13 had low and high suicide risks, respectively. The structured interview guide for the Hamilton Depression Rating Scale (HAMD) was adopted to assess suicide risk and depression severity. Suicide risk was clinician rated based on a suicide-related question (H11). The interviews were transcribed and the words in participants' verbal responses were translated into psychologically meaningful categories using Linguistic Inquiry and Word Count (LIWC). Results: Ordinal logistic regression revealed significant suicide-related language features in participants' responses to the HAMD questions. Increased use of anger words when talking about work and activities posed the highest suicide risk (odds ratio [OR] 2.91, 95% CI 1.22-8.55; P=.02). Random forest models demonstrated that text analysis of the direct responses to H11 was effective in identifying individuals with high suicide risk (AUC 0.76-0.89; P<.001) and detecting suicide risk in general, including both low and high suicide risk (AUC 0.83-0.92; P<.001). More importantly, suicide risk can be detected with satisfactory performance even without patients' disclosure of suicidal ideation. Based on the response to the question on hypochondriasis, ML models were trained to identify individuals with high suicide risk (AUC 0.76; P<.001). Conclusions: This study examined the perspective of using NLP and ML to analyze the texts from clinical interviews for suicidality detection, which has the potential to provide more accurate and specific markers for suicidal ideation detection. The findings may pave the way for developing high-performance assessment of suicide risk for automated detection, including online chatbot-based interviews for universal screening.

Deciphering associations between gut microbiota and clinical factors using microbial modules.

MOTIVATION: Human gut microbiota plays a vital role in maintaining body health. The dysbiosis of gut microbiota is associated with a variety of diseases. It is critical to uncover the associations between gut microbiota and disease states as well as other intrinsic or environmental factors. However, inferring alterations of individual microbial taxa based on relative abundance data likely leads to false associations and conflicting discoveries in different studies. Moreover, the effects of underlying factors and microbe-microbe interactions could lead to the alteration of larger sets of taxa. It might be more robust to investigate gut microbiota using groups of related taxa instead of the composition of individual taxa. RESULTS: We proposed a novel method to identify underlying microbial modules, i.e. groups of taxa with similar abundance patterns affected by a common latent factor, from longitudinal gut microbiota and applied it to inflammatory bowel disease (IBD). The identified modules demonstrated closer intragroup relationships, indicating potential microbe-microbe interactions and influences of underlying factors. Associations between the modules and several clinical factors were investigated, especially disease states. The IBD-associated modules performed better in stratifying the subjects compared with the relative abundance of individual taxa. The modules were further validated in external cohorts, demonstrating the efficacy of the proposed method in identifying general and robust microbial modules. The study reveals the benefit of considering the ecological effects in gut microbiota analysis and the great promise of linking clinical factors with underlying microbial modules. AVAILABILITY AND IMPLEMENTATION: https://github.com/rwang-z/microbial_module.git.

bvnGPS: a generalizable diagnostic model for acute bacterial and viral infection using integrative host transcriptomics and pretrained neural networks.

MOTIVATION: The confusion of acute inflammation infected by virus and bacteria or noninfectious inflammation will lead to missing the best therapy occasion resulting in poor prognoses. The diagnostic model based on host gene expression has been widely used to diagnose acute infections, but the clinical usage was hindered by the capability across different samples and cohorts due to the small sample size for signature training and discovery. RESULTS: Here, we construct a large-scale dataset integrating multiple host transcriptomic data and analyze it using a sophisticated strategy which removes batch effect and extracts the common information from different cohorts based on the relative expression alteration of gene pairs. We assemble 2680 samples across 16 cohorts and separately build gene pair signature (GPS) for bacterial, viral, and noninfected patients. The three GPSs are further assembled into an antibiotic decision model (bacterial-viral-noninfected GPS, bvnGPS) using multiclass neural networks, which is able to determine whether a patient is bacterial infected, viral infected, or noninfected. bvnGPS can distinguish bacterial infection with area under the receiver operating characteristic curve (AUC) of 0.953 (95% confidence interval, 0.948-0.958) and viral infection with AUC of 0.956 (0.951-0.961) in the test set (N = 760). In the validation set (N = 147), bvnGPS also shows strong performance by attaining an AUC of 0.988 (0.978-0.998) on bacterial-versus-other and an AUC of 0.994 (0.984-1.000) on viral-versus-other. bvnGPS has the potential to be used in clinical practice and the proposed procedure provides insight into data integration, feature selection and multiclass classification for host transcriptomics data. AVAILABILITY AND IMPLEMENTATION: The codes implementing bvnGPS are available at https://github.com/Ritchiegit/bvnGPS. The construction of iPAGE algorithm and the training of neural network was conducted on Python 3.7 with Scikit-learn 0.24.1 and PyTorch 1.7. The visualization of the results was implemented on R 4.2, Python 3.7, and Matplotlib 3.3.4.

Whole transcriptome analysis reveals non-coding RNA's competing endogenous gene pairs as novel form of motifs in serous ovarian cancer.

The non-coding RNA (ncRNA) regulation appears to be associated to the diagnosis and targeted therapy of complex diseases. Motifs of non-coding RNAs and genes in the competing endogenous RNA (ceRNA) network would probably contribute to the accurate prediction of serous ovarian carcinoma (SOC). We conducted a microarray study profiling the whole transcriptomes of eight human SOCs and eight controls and constructed a ceRNA network including mRNAs, long ncRNAs, and circular RNAs (circRNAs). Novel form of motifs (mRNA-ncRNA-mRNA) were identified from the ceRNA network and defined as non-coding RNA's competing endogenous gene pairs (ceGPs), using a proposed method denoised individualized pair analysis of gene expression (deiPAGE). 18 cricRNA's ceGPs (cceGPs) were identified from multiple cohorts and were fused as an indicator (SOC index) for SOC discrimination, which carried a high predictive capacity in independent cohorts. SOC index was negatively correlated with the CD8+/CD4+ ratio in tumour-infiltration, reflecting the migration and growth of tumour cells in ovarian cancer progression. Moreover, most of the RNAs in SOC index were experimentally validated involved in ovarian cancer development. Our results elucidate the discriminative capability of SOC index and suggest that the novel competing endogenous motifs play important roles in expression regulation and could be potential target for investigating ovarian cancer mechanism or its therapy.

meGPS: a multi-omics signature for hepatocellular carcinoma detection integrating methylome and transcriptome data.

MOTIVATION: Hepatocellular carcinoma (HCC) is a primary malignancy with a poor prognosis. Recently, multi-omics molecular-level measurement enables HCC diagnosis and prognosis prediction, which is crucial for early intervention of personalized therapy to diminish mortality. Here, we introduce a novel strategy utilizing DNA methylation and RNA expression data to achieve a multi-omics gene pair signature (GPS) for HCC discrimination. RESULTS: The immune genes with negative correlations between expression and promoter methylation are enriched in the highly connected cancer-related pathway network, which are considered as the candidates for HCC detection. After that, we separately construct a methylation GPS (mGPS) and an expression GPS (eGPS), and then assemble them as a meGPS with five gene pairs, in which the significant methylation and expression changes occur between HCC tumor and non-tumor groups. Reliable performance has been validated by independent tissue (age, gender and etiology) and blood datasets. This study proposes a procedure for multi-omics GPS identification and develops a novel HCC signature using both methylome and transcriptome data, suggesting potential molecular targets for the detection and therapy of HCC. AVAILABILITY AND IMPLEMENTATION: Models are available at https://github.com/bioinformaticStudy/meGPS.git. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Improving bulk RNA-seq classification by transferring gene signature from single cells in acute myeloid leukemia.

The advances in single-cell RNA sequencing (scRNA-seq) technologies enable the characterization of transcriptomic profiles at the cellular level and demonstrate great promise in bulk sample analysis thereby offering opportunities to transfer gene signature from scRNA-seq to bulk data. However, the gene expression signatures identified from single cells are typically inapplicable to bulk RNA-seq data due to the profiling differences of distinct sequencing technologies. Here, we propose single-cell pair-wise gene expression (scPAGE), a novel method to develop single-cell gene pair signatures (scGPSs) that were beneficial to bulk RNA-seq classification to transfer knowledge across platforms. PAGE was adopted to tackle the challenge of profiling differences. We applied the method to acute myeloid leukemia (AML) and identified the scGPS from mouse scRNA-seq that allowed discriminating between AML and control cells. The scGPS was validated in bulk RNA-seq datasets and demonstrated better performance (average area under the curve [AUC] = 0.96) than the conventional gene expression strategies (average AUC$\le$ 0.88) suggesting its potential in disclosing the molecular mechanism of AML. The scGPS also outperformed its bulk counterpart, which highlighted the benefit of gene signature transfer. Furthermore, we confirmed the utility of scPAGE in sepsis as an example of other disease scenarios. scPAGE leveraged the advantages of single-cell profiles to enhance the analysis of bulk samples revealing great potential of transferring knowledge from single-cell to bulk transcriptome studies.

A Robust and Generalizable Immune-Related Signature for Sepsis Diagnostics.

High-throughput sequencing can detect tens of thousands of genes in parallel, providing opportunities for improving the diagnostic accuracy of multiple diseases including sepsis, which is an aggressive inflammatory response to infection that can cause organ failure and death. Early screening of sepsis is essential in clinic, but no effective diagnostic biomarkers are available yet. Here, we present a novel method, Recurrent Logistic Regression, to identify diagnostic biomarkers for sepsis from the blood transcriptome data. A panel including five immune-related genes, LRRN3, IL2RB, FCER1A, TLR5, and S100A12, are determined as diagnostic biomarkers (LIFTS) for sepsis. LIFTS discriminates patients with sepsis from normal controls in high accuracy (AUROC = 0.9959 on average; IC = [0.9722-1.0]) on nine validation cohorts across three independent platforms, which outperforms existing markers. Our analysis determined an accurate prediction model and reproducible transcriptome biomarkers that can lay a foundation for clinical diagnostic tests and biological mechanistic studies.

Discovery of a New CDK4/6 and PI3K/AKT Multiple Kinase Inhibitor Aminoquinol for the Treatment of Hepatocellular Carcinoma.

Background: Hepatocellular carcinoma (HCC) is a lethal malignancy lacking effective treatment. The Cyclin-dependent kinases 4/6 (CDK4/6) and PI3K/AKT signal pathways play pivotal roles in carcinogenesis and are promising therapeutic targets for HCC. Here we identified a new CDK4/6 and PI3K/AKT multi-kinase inhibitor for the treatment of HCC. Methods: Using a repurposing and ensemble docking methodology, we screened a library of worldwide approved drugs to identify candidate CDK4/6 inhibitors. By MTT, apoptosis, and flow cytometry analysis, we investigated the effects of candidate drug in reducing cell-viability,inducing apoptosis, and causing cell-cycle arrest. The drug combination and thermal proteomic profiling (TPP) method were used to investigate whether the candidate drug produced antagonistic effect. The in vivo anti-cancer effect was performed in BALB/C nude mice subcutaneously xenografted with Huh7 cells. Results: We demonstrated for the first time that the anti-plasmodium drug aminoquinol is a new CDK4/6 and PI3K/AKT inhibitor. Aminoquinol significantly decreased cell viability, induced apoptosis, increased the percentage of cells in G1 phase. Drug combination screening indicated that aminoquinol could produce antagonistic effect with the PI3K inhibitor LY294002. TPP analysis confirmed that aminoquinol significantly stabilized CDK4, CDK6, PI3K and AKT proteins. Finally, in vivo study in Huh7 cells xenografted nude mice demonstrated that aminoquinol exhibited strong anti-tumor activity, comparable to that of the leading cancer drug 5-fluorouracil with the combination treatment showed the highest therapeutic effect. Conclusion: The present study indicates for the first time the discovery of a new CDK4/6 and PI3K/AKT multi-kinase inhibitor aminoquinol. It could be used alone or as a combination therapeutic strategy for the treatment of HCC.

Direct Measurement of B Lymphocyte Gene Expression Biomarkers in Peripheral Blood Transcriptomics Enables Early Prediction of Vaccine Seroconversion.

Peripheral blood transcriptome is a highly promising area for biomarker development. However, transcript abundances (TA) in these cell mixture samples are confounded by proportions of the component leukocyte subpopulations. This poses a challenge to clinical applications, as the cell of origin of any change in TA is not known without prior cell separation procedure. We developed a framework to develop a cell-type informative TA biomarkers which enable determination of TA of a single cell-type (B lymphocytes) directly in cell mixture samples of peripheral blood (e.g., peripheral blood mononuclear cells, PBMC) without the need for subpopulation separation. It is applicable to a panel of genes called B cell informative genes. Then a ratio of two B cell informative genes (a target gene and a stably expressed reference gene) obtained in PBMC was used as a new biomarker to represent the target gene expression in purified B lymphocytes. This approach, which eliminates the tedious procedure of cell separation and directly determines TA of a leukocyte subpopulation in peripheral blood samples, is called the Direct LS-TA method. This method is applied to gene expression datasets collected in influenza vaccination trials as early predictive biomarkers of seroconversion. By using TNFRSF17 or TXNDC5 as the target genes and TNFRSF13C or FCRLA as the reference genes, the Direct LS-TA B cell biomarkers were determined directly in the PBMC transcriptome data and were highly correlated with TA of the corresponding target genes in purified B lymphocytes. Vaccination responders had almost a 2-fold higher Direct LS-TA biomarker level of TNFRSF17 (log 2 SMD = 0.84, 95% CI = 0.47-1.21) on day 7 after vaccination. The sensitivity of these Direct LS-TA biomarkers in the prediction of seroconversion was greater than 0.7 and area-under curves (AUC) were over 0.8 in many datasets. In this paper, we report a straightforward approach to directly estimate B lymphocyte gene expression in PBMC, which could be used in a routine clinical setting. Moreover, the method enables the practice of precision medicine in the prediction of vaccination response. More importantly, seroconversion could now be predicted as early as day 7. As the acquired immunology pathway is common to vaccination against influenza and COVID-19, these biomarkers could also be useful to predict seroconversion for the new COVID-19 vaccines.

Machine-learning scoring functions trained on complexes dissimilar to the test set already outperform classical counterparts on a blind benchmark.

The superior performance of machine-learning scoring functions for docking has caused a series of debates on whether it is due to learning knowledge from training data that are similar in some sense to the test data. With a systematically revised methodology and a blind benchmark realistically mimicking the process of prospective prediction of binding affinity, we have evaluated three broadly used classical scoring functions and five machine-learning counterparts calibrated with both random forest and extreme gradient boosting using both solo and hybrid features, showing for the first time that machine-learning scoring functions trained exclusively on a proportion of as low as 8% complexes dissimilar to the test set already outperform classical scoring functions, a percentage that is far lower than what has been recently reported on all the three CASF benchmarks. The performance of machine-learning scoring functions is underestimated due to the absence of similar samples in some artificially created training sets that discard the full spectrum of complexes to be found in a prospective environment. Given the inevitability of any degree of similarity contained in a large dataset, the criteria for scoring function selection depend on which one can make the best use of all available materials. Software code and data are provided at https://github.com/cusdulab/MLSF for interested readers to rapidly rebuild the scoring functions and reproduce our results, even to make extended analyses on their own benchmarks.

Evaluating the Consistency of Gene Methylation in Liver Cancer Using Bisulfite Sequencing Data.

Bisulfite sequencing is considered as the gold standard approach for measuring DNA methylation, which acts as a pivotal part in regulating a variety of biological processes without changes in DNA sequences. In this study, we introduced the most prevalent methods for processing bisulfite sequencing data and evaluated the consistency of the data acquired from different measurements in liver cancer. Firstly, we introduced three commonly used bisulfite sequencing assays, i.e., reduced-representation bisulfite sequencing (RRBS), whole-genome bisulfite sequencing (WGBS), and targeted bisulfite sequencing (targeted BS). Next, we discussed the principles and compared different methods for alignment, quality assessment, methylation level scoring, and differentially methylated region identification. After that, we screened differential methylated genes in liver cancer through the three bisulfite sequencing assays and evaluated the consistency of their results. Ultimately, we compared bisulfite sequencing to 450 k beadchip and assessed the statistical similarity and functional association of differentially methylated genes (DMGs) among the four assays. Our results demonstrated that the DMGs measured by WGBS, RRBS, targeted BS and 450 k beadchip are consistently hypo-methylated in liver cancer with high functional similarity.

Long non-coding RNA pairs to assist in diagnosing sepsis.

BACKGROUND: Sepsis is the major cause of death in Intensive Care Unit (ICU) globally. Molecular detection enables rapid diagnosis that allows early intervention to minimize the death rate. Recent studies showed that long non-coding RNAs (lncRNAs) regulate proinflammatory genes and are related to the dysfunction of organs in sepsis. Identifying lncRNA signature with absolute abundance is challenging because of the technical variation and the systematic experimental bias. RESULTS: Cohorts (n = 768) containing whole blood lncRNA profiling of sepsis patients in the Gene Expression Omnibus (GEO) database were included. We proposed a novel diagnostic strategy that made use of the relative expressions of lncRNA pairs, which are reversed between sepsis patients and normal controls (eg. lncRNAi > lncRNAj in sepsis patients and lncRNAi < lncRNAj in normal controls), to identify 14 lncRNA pairs as a sepsis diagnostic signature. The signature was then applied to independent cohorts (n = 644) to evaluate its predictive performance across different ages and normalization methods. Comparing to common machine learning models and existing signatures, SepSigLnc consistently attains better performance on the validation cohorts from the same age group (AUC = 0.990 & 0.995 in two cohorts) and across different groups (AUC = 0.878 on average), as well as cohorts processed by an alternative normalization method (AUC = 0.953 on average). Functional analysis demonstrates that the lncRNA pairs in SepsigLnc are functionally similar and tend to implicate in the same biological processes including cell fate commitment and cellular response to steroid hormone stimulus. CONCLUSION: Our study identified 14 lncRNA pairs as signature that can facilitate the diagnosis of septic patients at an intervenable point when clinical manifestations are not dramatic. Also, the computational procedure can be generalized to a standard procedure for discovering diagnostic molecule signatures.

Discovery of vanoxerine dihydrochloride as a CDK2/4/6 triple-inhibitor for the treatment of human hepatocellular carcinoma.

BACKGROUND: Cyclin-dependent kinases 2/4/6 (CDK2/4/6) play critical roles in cell cycle progression, and their deregulations are hallmarks of hepatocellular carcinoma (HCC). METHODS: We used the combination of computational and experimental approaches to discover a CDK2/4/6 triple-inhibitor from FDA approved small-molecule drugs for the treatment of HCC. RESULTS: We identified vanoxerine dihydrochloride as a new CDK2/4/6 inhibitor, and a strong cytotoxicdrugin human HCC QGY7703 and Huh7 cells (IC50: 3.79 µM for QGY7703and 4.04 µM for Huh7 cells). In QGY7703 and Huh7 cells, vanoxerine dihydrochloride treatment caused G1-arrest, induced apoptosis, and reduced the expressions of CDK2/4/6, cyclin D/E, retinoblastoma protein (Rb), as well as the phosphorylation of CDK2/4/6 and Rb. Drug combination study indicated that vanoxerine dihydrochloride and 5-Fu produced synergistic cytotoxicity in vitro in Huh7 cells. Finally, in vivo study in BALB/C nude mice subcutaneously xenografted with Huh7 cells, vanoxerine dihydrochloride (40 mg/kg, i.p.) injection for 21 days produced significant anti-tumor activity (p < 0.05), which was comparable to that achieved by 5-Fu (10 mg/kg, i.p.), with the combination treatment resulted in synergistic effect. Immunohistochemistry staining of the tumor tissues also revealed significantly reduced expressions of Rb and CDK2/4/6in vanoxerinedihydrochloride treatment group. CONCLUSIONS: The present study isthe first report identifying a new CDK2/4/6 triple inhibitor vanoxerine dihydrochloride, and demonstrated that this drug represents a novel therapeutic strategy for HCC treatment.

A network-based algorithm for the identification of moonlighting noncoding RNAs and its application in sepsis.

Moonlighting proteins provide more options for cells to execute multiple functions without increasing the genome and transcriptome complexity. Although there have long been calls for computational methods for the prediction of moonlighting proteins, no method has been designed for determining moonlighting long noncoding ribonucleicacidz (RNAs) (mlncRNAs). Previously, we developed an algorithm MoonFinder for the identification of mlncRNAs at the genome level based on the functional annotation and interactome data of lncRNAs and proteins. Here, we update MoonFinder to MoonFinder v2.0 by providing an extensive framework for the detection of protein modules and the establishment of RNA-module associations in human. A novel measure, moonlighting coefficient, was also proposed to assess the confidence of an ncRNA acting in a moonlighting manner. Moreover, we explored the expression characteristics of mlncRNAs in sepsis, in which we found that mlncRNAs tend to be upregulated and differentially expressed. Interestingly, the mlncRNAs are mutually exclusive in terms of coexpression when compared to the other lncRNAs. Overall, MoonFinder v2.0 is dedicated to the prediction of human mlncRNAs and thus bears great promise to serve as a valuable R package for worldwide research communities (https://cran.r-project.org/web/packages/MoonFinder/index.html). Also, our analyses provide the first attempt to characterize mlncRNA expression and coexpression properties in adult sepsis patients, which will facilitate the understanding of the interaction and expression patterns of mlncRNAs.

Probabilistic Contextual and Structural Dependencies Learning in Grammar-Based Genetic Programming.

Genetic Programming is a method to automatically create computer programs based on the principles of evolution. The problem of deceptiveness caused by complex dependencies among components of programs is challenging. It is important because it can misguide Genetic Programming to create suboptimal programs. Besides, a minor modification in the programs may lead to a notable change in the program behaviours and affect the final outputs. This article presents Grammar-Based Genetic Programming with Bayesian Classifiers (GBGPBC) in which the probabilistic dependencies among components of programs are captured using a set of Bayesian network classifiers. Our system was evaluated using a set of benchmark problems (the deceptive maximum problems, the royal tree problems, and the bipolar asymmetric royal tree problems). It was shown to be often more robust and more efficient in searching the best programs than other related Genetic Programming approaches in terms of the total number of fitness evaluation. We studied what factors affect the performance of GBGPBC and discovered that robust variants of GBGPBC were consistently weakly correlated with some complexity measures. Furthermore, our approach has been applied to learn a ranking program on a set of customers in direct marketing. Our suggested solutions help companies to earn significantly more when compared with other solutions produced by several well-known machine learning algorithms, such as neural networks, logistic regression, and Bayesian networks.

Systematic prediction of autophagy-related proteins using Arabidopsis thaliana interactome data.

Autophagy is a self-degradative process that is crucial for maintaining cellular homeostasis by removing damaged cytoplasmic components and recycling nutrients. Such an evolutionary conserved proteolysis process is regulated by the autophagy-related (Atg) proteins. The incomplete understanding of plant autophagy proteome and the importance of a proteome-wide understanding of the autophagy pathway prompted us to predict Atg proteins and regulators in Arabidopsis. Here, we developed a systems-level algorithm to identify autophagy-related modules (ARMs) based on protein subcellular localization, protein-protein interactions, and known Atg proteins. This generates a detailed landscape of the autophagic modules in Arabidopsis. We found that the newly identified genes in each ARM tend to be upregulated and coexpressed during the senescence stage of Arabidopsis. We also demonstrated that the Golgi apparatus ARM, ARM13, functions in the autophagy process by module clustering and functional analysis. To verify the in silico analysis, the Atg candidates in ARM13 that are functionally similar to the core Atg proteins were selected for experimental validation. Interestingly, two of the previously uncharacterized proteins identified from the ARM analysis, AGD1 and Sec14, exhibited bona fide association with the autophagy protein complex in plant cells, which provides evidence for a cross-talk between intracellular pathways and autophagy. Thus, the computational framework has facilitated the identification and characterization of plant-specific autophagy-related proteins and novel autophagy proteins/regulators in higher eukaryotes.

Genome-Wide Search for SNP Interactions in GWAS Data: Algorithm, Feasibility, Replication Using Schizophrenia Datasets.

In this study, we looked for potential gene-gene interaction in susceptibility to schizophrenia by an exhaustive searching for SNP-SNP interactions in 3 GWAS datasets (phs000021:phg000013, phs000021:phg000014, phs000167) using our recently published algorithm. The search space for SNP-SNP interaction was confined to 8 biologically plausible ways of interaction under dominant-dominant or recessive-recessive modes. First, we performed our search of all pair-wise combination of 729,454 SNPs after filtering by SNP genotype quality. All possible pairwise interactions of any 2 SNPs (5 × 1011) were exhausted to search for significant interaction which was defined by p-value of chi-square tests. Nine out the top 10 interactions, protein coding genes were partnered with non-coding RNA (ncRNA) which suggested a new alternative insight into interaction biology other than the frequently sought-after protein-protein interaction. Therefore, we extended to look for replication among the top 10,000 interaction SNP pairs and high proportion of concurrent genes forming the interaction pairs were found. The results indicated that an enrichment of signals over noise was present in the top 10,000 interactions. Then, replications of SNP-SNP interaction were confirmed for 14 SNPs-pairs in both replication datasets. Biological insight was highlighted by a potential binding between FHIT (protein coding gene) and LINC00969 (lncRNA) which showed a replicable interaction between their SNPs. Both of them were reported to have expression in brain. Our study represented an early attempt of exhaustive interaction analysis of GWAS data which also yield replicated interaction and new insight into understanding of genetic interaction in schizophrenia.

Predicting associations among drugs, targets and diseases by tensor decomposition for drug repositioning.

BACKGROUND: Development of new drugs is a time-consuming and costly process, and the cost is still increasing in recent years. However, the number of drugs approved by FDA every year per dollar spent on development is declining. Drug repositioning, which aims to find new use of existing drugs, attracts attention of pharmaceutical researchers due to its high efficiency. A variety of computational methods for drug repositioning have been proposed based on machine learning approaches, network-based approaches, matrix decomposition approaches, etc. RESULTS: We propose a novel computational method for drug repositioning. We construct and decompose three-dimensional tensors, which consist of the associations among drugs, targets and diseases, to derive latent factors reflecting the functional patterns of the three kinds of entities. The proposed method outperforms several baseline methods in recovering missing associations. Most of the top predictions are validated by literature search and computational docking. Latent factors are used to cluster the drugs, targets and diseases into functional groups. Topological Data Analysis (TDA) is applied to investigate the properties of the clusters. We find that the latent factors are able to capture the functional patterns and underlying molecular mechanisms of drugs, targets and diseases. In addition, we focus on repurposing drugs for cancer and discover not only new therapeutic use but also adverse effects of the drugs. In the in-depth study of associations among the clusters of drugs, targets and cancer subtypes, we find there exist strong associations between particular clusters. CONCLUSIONS: The proposed method is able to recover missing associations, discover new predictions and uncover functional clusters of drugs, targets and diseases. The clustering of drugs, targets and diseases, as well as the associations among the clusters, provides a new guiding framework for drug repositioning.