Search

Unsupervised feature selection is a critical step for efficient and accurate analysis of single-cell RNA-seq data. Previous benchmarks used two different criteria to compare feature selection methods: (i) proportion of ground-truth marker genes included in the selected features and (ii) accuracy of cell clustering using ground-truth cell types. Here, we systematically compare the performance of 11 feature selection methods for both criteria. We first demonstrate the discordance between these criteria and suggest using the latter. We then compare the distribution of selected genes in their means between feature selection methods. We show that lowly expressed genes exhibit seriously high coefficients of variation and are mostly excluded by high-performance methods. In particular, high-deviation- and high-expression-based methods outperform the widely used in Seurat package in clustering cells and data visualization. We further show they also enable a clear separation of the same cell type from different tissues as well as accurate estimation of cell trajectories. [ABSTRACT FROM AUTHOR]

Abstract Background Leishmaniasis is a virulent parasitic infection that causes a worldwide disease burden. Most treatments have toxic side-effects and efficacy has decreased due to the emergence of resistant strains. The outlook is worsened by the absence of promising drug targets for this disease. We have taken a computational approach to the detection of new drug targets, which may become an effective strategy for the discovery of new drugs for this tropical disease. Results We have predicted the protein interaction network of Leishmania major by using three validated methods: PSIMAP, PEIMAP, and iPfam. Combining the results from these methods, we calculated a high confidence network (confidence score > 0.70) with 1,366 nodes and 33,861 interactions. We were able to predict the biological process for 263 interacting proteins by doing enrichment analysis of the clusters detected. Analyzing the topology of the network with metrics such as connectivity and betweenness centrality, we detected 142 potential drug targets after homology filtering with the human proteome. Further experiments can be done to validate these targets. Conclusion We have constructed the first protein interaction network of the Leishmania major parasite by using a computational approach. The topological analysis of the protein network enabled us to identify a set of candidate proteins that may be both (1) essential for parasite survival and (2) without human orthologs. These potential targets are promising for further experimental validation. This strategy, if validated, may augment established drug discovery methodologies, for this and possibly other tropical diseases, with a relatively low additional investment of time and resources.

Thirteen compounds were isolated from the Canavalia lineata pods and their inhibitory activities against human monoamine oxidase-A (hMAO-A) and -B (hMAO-B) were evaluated. Among them, compounds 8 (medicarpin) and 13 (homopterocarpin) showed potent inhibitory activity against hMAO-B (IC50 = 0.45 and 0.72 µM, respectively) with selectivity index (SI) values of 44.2 and 2.07, respectively. Most of the compounds weakly inhibited MAO-A, except 9 (prunetin) and 13. Compounds 8 and 13 were reversible competitive inhibitors against hMAO-B (Ki = 0.27 and 0.21 µM, respectively). Structurally, the 3-OH group at A-ring of 8 showed higher hMAO-B inhibitory activity than 3-OCH3 group at the A-ring of 13. However, the 9-OCH3 group at B-ring of 13 showed higher hMAO-B inhibitory activity than 8,9-methylenedioxygroup at the B-ring of 12 (pterocarpin). In cytotoxicity study, 8 and 13 showed non-toxicity to the normal (MDCK) and cancer (HL-60) cells and moderate toxicity to neuroblastoma (SH-SY5Y) cell. Molecular docking simulation revealed that the binding affinities of 8 and 13 for hMAO-B (−8.7 and −7.7 kcal/mol, respectively) were higher than those for hMAO-A (−3.4 and −7.1 kcal/mol, respectively). These findings suggest that compounds 8 and 13 be considered potent reversible hMAO-B inhibitors to be used for the treatment of neurological disorders. [ABSTRACT FROM AUTHOR]

Abstract Background A disease-causing mutation refers to a heritable genetic change that is associated with a specific phenotype (disease). The detection of a mutation from a patient's sample is critical for the diagnosis, treatment, and prognosis of the disease. There are numerous databases and applications with which to archive mutation data. However, none of them have been implemented with any automated bioinformatics tools for mutation detection and analysis starting from raw data materials from patients. We present a Locus Specific mutation DB (LSDB) construction system that supports both mutation detection and deposition in one package. Results COMUS (Clinician-Oriented locus specific MUtation detection and deposition System) is a mutation detection and deposition system for developing specific LSDBs. COMUS contains 1) a DNA sequence mutation analysis method for clinicians' mutation data identification and deposition and 2) a curation system for variation detection from clinicians' input data. To embody the COMUS system and to validate its clinical utility, we have chosen the disease hemophilia as a test database. A set of data files from bench experiments and clinical information from hemophilia patients were tested on the LSDB, KoHemGene http://www.kohemgene.org, which has proven to be a clinician-friendly interface for mutation detection and deposition. Conclusion COMUS is a bioinformatics system for detecting and depositing new mutations from patient DNA with a clinician-friendly interface. LSDBs made using COMUS will promote the clinical utility of LSDBs. COMUS is available at http://www.comus.info.

Abstract Background Pseudomonas putida KT2440 (P. putida KT2440) is a highly versatile saprophytic soil bacterium. It is a certified bio-safety host for transferring foreign genes. Therefore, the bacterium is used as a model organism for genetic and physiological studies and for the development of biotechnological applications. In order to provide a more systematic application of the organism, we have constructed a protein-protein interaction (PPI) network analysis system of P. putida KT2440. Results PutidaNET is a comprehensive interaction database and server of P. putida KT2440 which is generated from three protein-protein interaction (PPI) methods. We used PSIMAP (Protein Structural Interactome MAP), PEIMAP (Protein Experimental Interactome MAP), and Domain-domain interactions using iPfam. PutidaNET contains 3,254 proteins, and 82,019 possible interactions consisting of 61,011 (PSIMAP), 4,293 (PEIMAP), and 30,043 (iPfam) interaction pairs except for self interaction. Also, we performed a case study by integrating a protein interaction network and experimental 1-DE/MS-MS analysis data P. putida. We found that 1) major functional modules are involved in various metabolic pathways and ribosomes, and 2) existing PPI sub-networks that are specific to succinate or benzoate metabolism are not in the center as predicted. Conclusion We introduce the PutidaNET which provides predicted interaction partners and functional analyses such as physicochemical properties, KEGG pathway assignment, and Gene Ontology mapping of P. putida KT2440 PutidaNET is freely available at http://sequenceome.kobic.kr/PutidaNET.

Abstract Background Mitochondria play a vital role in the energy production and apoptotic process of eukaryotic cells. Proteins in the mitochondria are encoded by nuclear and mitochondrial genes. Owing to a large increase in the number of identified mitochondrial protein sequences and completed mitochondrial genomes, it has become necessary to provide a web-based database of mitochondrial protein information. Results We present 'MitoInteractome', a consolidated web-based portal containing a wealth of information on predicted protein-protein interactions, physico-chemical properties, polymorphism, and diseases related to the mitochondrial proteome. MitoInteractome contains 6,549 protein sequences which were extracted from the following databases: SwissProt, MitoP, MitoProteome, HPRD and Gene Ontology database. The first general mitochondrial interactome has been constructed based on the concept of 'homologous interaction' using PSIMAP (Protein Structural Interactome MAP) and PEIMAP (Protein Experimental Interactome MAP). Using the above mentioned methods, protein-protein interactions were predicted for 74 species. The mitochondrial protein interaction data of humans was used to construct a network for the aging process. Analysis of the 'aging network' gave us vital insights into the interactions among proteins that influence the aging process. Conclusion MitoInteractome is a comprehensive database that would (1) aid in increasing our understanding of the molecular functions and interaction networks of mitochondrial proteins, (2) help in identifying new target proteins for experimental research using predicted protein-protein interaction information, and (3) help in identifying biomarkers for diagnosis and new molecular targets for drug development related to mitochondria. MitoInteractome is available at http://mitointeractome.kobic.kr/.

Abstract Background Analysis of single nucleotide polymorphism (SNP) is becoming a key research in genomics fields. Many functional analyses of SNPs have been carried out for coding regions and splicing sites that can alter proteins and mRNA splicing. However, SNPs in non-coding regulatory regions can also influence important biological regulation. Presently, there are few databases for SNPs in non-coding regulatory regions. Description We identified 488,452 human SNPs in the putative promoter regions that extended from the +5000 bp to -500 bp region of the transcription start sites. Some SNPs occurring in transcription factor (TF) binding sites were also predicted (47,832 SNP; 9.8%). The result is stored in a database: SNP@promoter. Users can search the SNP@Promoter database using three entries: 1) by SNP identifier (rs number from dbSNP), 2) by gene (gene name, gene symbol, refSeq ID), and 3) by disease term. The SNP@Promoter database provides extensive genetic information and graphical views of queried terms. Conclusion We present the SNP@Promoter database. It was created in order to predict functional SNPs in putative promoter regions and predicted transcription factor binding sites. SNP@Promoter will help researchers to identify functional SNPs in non-coding regions.