The evaluation of drug-gene-disease interactions is key for the identification of drugs effective against disease. However, at present, drugs that are effective against genes that are critical for disease are difficult to identify. Following a disease-centric approach, there is a need to identify genes critical to disease function and find drugs that are effective against them. By contrast, following a drug-centric approach comprises identifying the genes targeted by drugs, and then the diseases in which the identified genes are critical. Both of these processes are complex. Using a gene-centric approach, whereby we identify genes that are effective against the disease and can be targeted by drugs, is much easier. However, how such sets of genes can be identified without specifying either the target diseases or drugs is not known. In this study, a novel artificial intelligence-based approach that employs unsupervised methods and identifies genes without specifying neither diseases nor drugs is presented. To evaluate its feasibility, we applied tensor decomposition (TD)-based unsupervised feature extraction (FE) to perform drug repositioning from protein-protein interactions (PPI) without any other information. Proteins selected by TD-based unsupervised FE include many genes related to cancers, as well as drugs that target the selected proteins. Thus, we were able to identify cancer drugs using only PPI. Because the selected proteins had more interactions, we replaced the selected proteins with hub proteins and found that hub proteins themselves could be used for drug repositioning. In contrast to hub proteins, which can only identify cancer drugs, TD-based unsupervised FE enables the identification of drugs for other diseases. In addition, TD-based unsupervised FE can be used to identify drugs that are effective in in vivo experiments, which is difficult when hub proteins are used. In conclusion, TD-based unsupervised FE is a useful tool for drug repositioning using only PPI without other information.

In this work, we extended the recently developed tensor decomposition (TD) based unsupervised feature extraction (FE) to a kernel-based method through a mathematical formulation. Subsequently, the kernel TD (KTD) based unsupervised FE was applied to two synthetic examples as well as real data sets, and the findings were compared with those obtained previously using the TD-based unsupervised FE approaches. The KTD-based unsupervised FE outperformed or performed comparably with the TD-based unsupervised FE in large p small n situations, which are situations involving a limited number of samples with many variables (observations). Nevertheless, the KTD-based unsupervised FE outperformed the TD-based unsupervised FE in non large p small n situations. In general, although the use of the kernel trick can help the TD-based unsupervised FE gain more variations, a wider range of problems may also be encountered. Considering the outperformance or comparable performance of the KTD-based unsupervised FE compared to the TD-based unsupervised FE when applied to large p small n problems, it is expected that the KTD-based unsupervised FE can be applied in the genomic science domain, which involves many large p small n problems, and, in which, the TD-based unsupervised FE approach has been effectively applied.

BACKGROUND: Although in silico drug discovery is necessary for drug development, two major strategies, a structure-based and ligand-based approach, have not been completely successful. Currently, the third approach, inference of drug candidates from gene expression profiles obtained from the cells treated with the compounds under study requires the use of a training dataset. Here, the purpose was to develop a new approach that does not require any pre-existing knowledge about the drug-protein interactions, but these interactions can be inferred by means of an integrated approach using gene expression profiles obtained from the cells treated with the analysed compounds and the existing data describing gene-gene interactions. RESULTS: In the present study, using tensor decomposition-based unsupervised feature extraction, which represents an extension of the recently proposed principal-component analysis-based feature extraction, gene sets and compounds with a significant dose-dependent activity were screened without any training datasets. Next, after these results were combined with the data showing perturbations in single-gene expression profiles, genes targeted by the analysed compounds were inferred. The set of target genes thus identified was shown to significantly overlap with known target genes of the compounds under study. CONCLUSIONS: The method is specifically designed for large-scale datasets (including hundreds of treatments with compounds), not for conventional small-scale datasets. The obtained results indicate that two compounds that have not been extensively studied, WZ-3105 and CGP-60474, represent promising drug candidates targeting multiple cancers, including melanoma, adenocarcinoma, liver carcinoma, and breast, colon, and prostate cancers, which were analysed in this in silico study.

Integrated analysis of epigenetic profiles is important but difficult. Tensor decomposition-based unsupervised feature extraction was applied here to data on microRNA (miRNA) expression and promoter methylation of genes in ovarian cancer. It selected seven miRNAs and 241 genes by expression levels and promoter methylation degrees, respectively, such that they showed differences between eight normal ovarian tissue samples and 569 tumor samples. The expression levels of the seven miRNAs and the degrees of promoter methylation of the 241 genes also correlated significantly. Conventional Student's t test-based feature selection failed to identify miRNAs and genes that have the above properties. On the other hand, biological evaluation of the seven identified miRNAs and 241 identified genes suggests that they are strongly related to cancer as expected.

BACKGROUND: Even though coexistence of multiple phenotypes sharing the same genomic background is interesting, it remains incompletely understood. Epigenomic profiles may represent key factors, with unknown contributions to the development of multiple phenotypes, and social-insect castes are a good model for elucidation of the underlying mechanisms. Nonetheless, previous studies have failed to identify genes associated with aberrant gene expression and methylation profiles because of the lack of suitable methodology that can address this problem properly. METHODS: A recently proposed principal component analysis (PCA)-based and tensor decomposition (TD)-based unsupervised feature extraction (FE) can solve this problem because these two approaches can deal with gene expression and methylation profiles even when a small number of samples is available. RESULTS: PCA-based and TD-based unsupervised FE methods were applied to the analysis of gene expression and methylation profiles in the brains of two social insects, Polistes canadensis and Dinoponera quadriceps. Genes associated with differential expression and methylation between castes were identified, and analysis of enrichment of Gene Ontology terms confirmed reliability of the obtained sets of genes from the biological standpoint. CONCLUSIONS: Biologically relevant genes, shown to be associated with significant differential gene expression and methylation between castes, were identified here for the first time. The identification of these genes may help understand the mechanisms underlying epigenetic control of development of multiple phenotypes under the same genomic conditions.

Because usually there are no normal control cell lines, cancer cell lines can be examined only in a comparison between treatment and no-treatment conditions. Thus, characterization of cancer cell lines by themselves is impossible. To address this problem, one-class differential expression (DE) analysis, which can evaluate samples without a reference, is proposed here using tensor decomposition (TD)-based unsupervised feature extraction (FE) extended from recently proposed principal component analysis-based unsupervised FE. This one-class DE analysis was applied to multi-omics datasets of 26 lung adenocarcinoma cell lines. Enrichment analysis of selected genes identified multiple biological terms or concepts including signal recognition particles and nonsense-mediated decay (Reactome, Gene Ontology [GO] biological process), cadherin, poly(A) RNA binding (GO molecular function), eukaryotic translation initiation factors (Reactome), aberrant histone protein expression (Reactome and Human Protein Atlas [HPA]), and 163 transcription factors including E2F, PAX5, ARNT, AHR, and CREB, all of which are known to be related to non-small cell lung cancer and are expected to function cooperatively in lung adenocarcinoma oncogenesis. These data not only indicate usefulness of one-class DE analysis using TD-based unsupervised FE but also point to new therapeutic targets in lung adenocarcinoma.

Wilms tumor is one of lethal child renal cancers, for which no known disease causing mechanisms exist. In this paper, we tried to identify possible disease causing microRNA(miRNA)-mRNA pairs (interactions) by analyzing (partially matched) miRNA/mRNA gene expression profiles with the recently proposed principal component analysis based unsupervised feature extraction. It successfully identified multiple miRNA-mRNA pairs whose biological natures are convincing. Correlation coefficients between miRNA and mRNA expression in matched parts of profiles turned out to be significantly negative. Constructed miRNA-mRNA network will be a key to understand Wilms tumor causing mechanisms.

We applied principal component analysis (PCA)-based unsupervised feature extraction (FE) to amyotrophic lateral sclerosis (ALS) gene expression profiles. ALS is a debilitating neurodegenerative disorder with no effective therapy. The relevant gene expression profiles contained a small number of samples (from a few to tens) with a large number of features (several tens of thousands). Although it is important to recognize critical genes from gene expression profiles, a small-sample-large-feature situation makes FE difficult. In PCA-based unsupervised FE, features rather than samples are embedded into a low dimensional space, and critical genes are identified as outliers that are supposed to obey group-oriented behavior. The 29 candidate genes identified as critical for ALS by this methodology turned out to be biologically feasible based on comparisons with numerous previous studies. Together, they formed a collected gene regulation/protein binding network within which the known, but not explicitly identified in this study, three ALS-causing genes, SOD1, TDP-43, and SETX, could be naturally placed/embedded. Among the 29 genes, the translated chemokine receptor CCR6 protein was considered to be a potential therapy target and its antagonists/agonists were identified using the in silico drug discovery software ChooseLD. The ten top-ranked antagonists/agonists shared structures with many compounds that were previously known to bind to various proteins.

BACKGROUND: Transgenerational epigenetics (TGE) are currently considered important in disease, but the mechanisms involved are not yet fully understood. TGE abnormalities expected to cause disease are likely to be initiated during development and to be mediated by aberrant gene expression associated with aberrant promoter methylation that is heritable between generations. However, because methylation is removed and then re-established during development, it is not easy to identify promoter methylation abnormalities by comparing normal lineages with those expected to exhibit TGE abnormalities. METHODS: This study applied the recently proposed principal component analysis (PCA)-based unsupervised feature extraction to previously reported and publically available gene expression/promoter methylation profiles of rat primordial germ cells, between E13 and E16 of the F3 generation vinclozolin lineage that are expected to exhibit TGE abnormalities, to identify multiple genes that exhibited aberrant gene expression/promoter methylation during development. RESULTS: The biological feasibility of the identified genes were tested via enrichment analyses of various biological concepts including pathway analysis, gene ontology terms and protein-protein interactions. All validations suggested superiority of the proposed method over three conventional and popular supervised methods that employed t test, limma and significance analysis of microarrays, respectively. The identified genes were globally related to tumors, the prostate, kidney, testis and the immune system and were previously reported to be related to various diseases caused by TGE. CONCLUSIONS: Among the genes reported by PCA-based unsupervised feature extraction, we propose that chemokine signaling pathways and leucine rich repeat proteins are key factors that initiate transgenerational epigenetic-mediated diseases, because multiple genes included in these two categories were identified in this study.

BACKGROUND: Feature extraction (FE) is difficult, particularly if there are more features than samples, as small sample numbers often result in biased outcomes or overfitting. Furthermore, multiple sample classes often complicate FE because evaluating performance, which is usual in supervised FE, is generally harder than the two-class problem. Developing sample classification independent unsupervised methods would solve many of these problems. RESULTS: Two principal component analysis (PCA)-based FE, specifically, variational Bayes PCA (VBPCA) was extended to perform unsupervised FE, and together with conventional PCA (CPCA)-based unsupervised FE, were tested as sample classification independent unsupervised FE methods. VBPCA- and CPCA-based unsupervised FE both performed well when applied to simulated data, and a posttraumatic stress disorder (PTSD)-mediated heart disease data set that had multiple categorical class observations in mRNA/microRNA expression of stressed mouse heart. A critical set of PTSD miRNAs/mRNAs were identified that show aberrant expression between treatment and control samples, and significant, negative correlation with one another. Moreover, greater stability and biological feasibility than conventional supervised FE was also demonstrated. Based on the results obtained, in silico drug discovery was performed as translational validation of the methods. CONCLUSIONS: Our two proposed unsupervised FE methods (CPCA- and VBPCA-based) worked well on simulated data, and outperformed two conventional supervised FE methods on a real data set. Thus, these two methods have suggested equivalence for FE on categorical multiclass data sets, with potential translational utility for in silico drug discovery.

Usually prediction of molecular functions of proteins from their amino acid sequences is based upon sequence similarity with proteins of known functions. However, it is well known that function is mainly dependent upon protein structures than sequences. Since structures are often independent of sequences, it is important to predict function without sequence similarities. Here we propose a method based upon amino acid occurrence for predicting Gene Ontology (GO) term. We have tested the method in a set of 3212 proteins in Protein Data Bank with less than 40% sequence identity. Our method achieved more than 50% sensitivity and 20% precision for c.a. 20 selected GO terms among the most frequent 557 GO terms. Mean sensitivity, specificity, precision, and accuracy for relatively rare (but majority) 402 GO terms among the 557 GO terms are 13%, 99%, 9% and 99%, respectively. They are significantly larger than expected values of less than 2% under assuming random selection.

BACKGROUND: Predicting the three-dimensional structure of a protein from its amino acid sequence is a long-standing goal in computational/molecular biology. The discrimination of different structural classes and folding types are intermediate steps in protein structure prediction. RESULTS: In this work, we have proposed a method based on linear discriminant analysis (LDA) for discriminating 30 different folding types of globular proteins using amino acid occurrence. Our method was tested with a non-redundant set of 1612 proteins and it discriminated them with the accuracy of 38%, which is comparable to or better than other methods in the literature. A web server has been developed for discriminating the folding type of a query protein from its amino acid sequence and it is available at http://granular.com/PROLDA/. CONCLUSION: Amino acid occurrence has been successfully used to discriminate different folding types of globular proteins. The discrimination accuracy obtained with amino acid occurrence is better than that obtained with amino acid composition and/or amino acid properties. In addition, the method is very fast to obtain the results.

MOTIVATION: Microarray experiments result in large-scale data sets that require extensive mining and refining to extract useful information. We demonstrate the usefulness of (non-metric) multidimensional scaling (MDS) method in analyzing a large number of genes. Applying MDS to the microarray data is certainly not new, but the existing works are all on small numbers (< 100) of points to be analyzed. We have been developing an efficient novel algorithm for non-metric MDS (nMDS) analysis for very large data sets as a maximally unsupervised data mining device. We wish to demonstrate its usefulness in the context of bioinformatics (unraveling relational patterns among genes from time series data in this paper). RESULTS: The Pearson correlation coefficient with its sign flipped is used to measure the dissimilarity of the gene activities in transcriptional response of cell-cycle-synchronized human fibroblasts to serum. These dissimilarity data have been analyzed with our nMDS algorithm to produce an almost circular relational pattern of the genes. The obtained pattern expresses a temporal order in the data in this example; the temporal expression pattern of the genes rotates along this circular arrangement and is related to the cell cycle. For the data we analyze in this paper we observe the following. If an appropriate preparation procedure is applied to the original data set, linear methods such as the principal component analysis (PCA) could achieve reasonable results, but without data preprocessing linear methods such as PCA cannot achieve a useful picture. Furthermore, even with an appropriate data preprocessing, the outcomes of linear procedures are not as clear-cut as those by nMDS without preprocessing.