Skip to main content

Table 1 Summary of articles arranged by topic groups

From: A bioinformatics potpourri

Topic

Key finding or features

Algorithms

Divisive hierarchical maximum likelihood clustering (DRAGON), reduced search complexity O(n2c) [8].

2D–EM two-step clustering approach applied to transcriptome and methylome data; filtering produces a feature matrix which is used as input for clustering by modified EM [9].

Deep learning; convolutional neural network (CNN) using gradient boosted feature selection for classification of β-lactamases [10].

Bio-molecular networks

Prediction of protein complexes from protein-protein interaction networks utilizing gene expression data and protein functional annotations; CPredictor3.0 [11].

Boolean network model simulation of signaling networks; changes of modularity and robustness by edge-removal mutations [12].

Integrated protein-protein interaction network construction using DIP, Biogrid, Reactome and HPRD data; refinement/correction using relationship of functional similarity and proximity scores [13].

Boolean network modeling; topology comparison of gene-gene dynamics influence and gene-gene molecular interaction networks [14].

Integration of gene regulatory network inference with constraint-based metabolic models to simulate growth phenotype and exchange fluxes [15].

Causal relationship detection between gene pairs for short time-series gene expression data (E. coli; yeast) based on lagged-coordinate delay embedding theorem [16].

Cell fate predictions derived from polynomial modeling of human pancreatic single-cell gene expression data [17].

Case study of network bi-stability and positive/negative feedback loops in TGF-β1 activation [18].

Cancer and disease informatics

Multi-view clustering method with enhanced consensus; breast cancer sub-typing and survival analysis [19].

Network hubs as prognostic signatures in ovarian cancer, breast cancer and glioblastoma multiforme selected by Cox regression for correlating DNA methylation levels with outcome [20].

Analysis pipeline in Python to classify tumors using a supervised machine-learning algorithm that predicts mutation status based on transcriptional patterns [21].

Breast cancer outcome predictions from microarray data using Hadamard kernel [22].

Application of node-weighted Steiner tree approach to identify proteins and protein-protein interactions in PI3K/Akt and MAPK signaling pathways; subnetwork identification [23].

Tensor decomposition-based unsupervised feature extraction from gene expression data infers genes that induce post-traumatic stress disorder-mediated heart diseases and potential therapeutic targets [24].

Graph regression-based approach which creates a unified framework for predicting binary, discrete and continued lncRNA-disease associations [25].

Network consistency projection for human microbe-disease association predictions assuming that microbes with similar functions may have similar associated/not associated patterns with similar diseases [26].

Target control problem with objectives-guided optimization algorithm to identify drivers (e.g. drug target nodes or network biomarkers) controlling targets in disease-associated networks [27].

Drug-target interactions and drug efficacy

Ligand-based quantitative structure-activity relationship modeling using Random Forest for drug target identification; web server application [28].

Whole-body physiologically based pharmacokinetic modelling using constraint-based perturbation analysis with cluster Newton method; can handle mixed patient-dependent and patient-independent parameters [29].

Ternary status based integer linear programming approach for MCF7 breast cancer cell line specific pathway network reconstruction and prediction of treatment efficacy of compounds using prior knowledge of literature and phosphoproteomic data [30].

Core pharmacophore anchor model screening of FDA drugs to identify candidate dengue virus NS3 protease inhibitors [31].

Dependency-based deep neural network model for drug-drug interaction feature extraction form Drug Bank [32].

Gene expression and regulation

Prediction of human transcriptional target genes using open chromatin regions, ChIP-seq data and transcription factor binding sites [33].

Identification of phased secondary small interfering RNAs and miRNAs targeting PHAS loci in Panax notoginseng [34].

Identification candidate tissue-specific circRNAs using bi-clustering of RNA-Seq-derived expression profiles [35].

Multi-factor analysis of differential co-expression of breast cancer microarray data; identified differentially co-expressed sets containing ESR1 and CXCL13 [36].

Random Forest approach that uses motif combinations in prediction of cell-type-specific transcription factor binding sites [37].

Imaging

Discriminant filter bank approach for extracting EEG signal features using common spatial patterns; low misclassification rate [38].

MatQuantify is a software for assessment of fluorescence microscopy images of mitotic spindles and their architecture changes [39].

Phytoplankton and zooplankton image classification using a non-linear multiple kernel learning approach [40].

Immuno-informatics

Computational methodology pipeline to process, predict and analyze potential T cell epitopes using influenza A, dengue, West Nile hepatitis A and HIV-1 virus sequence data [41].

NetCTL-bases predictions of HLA-A2, -A3 and -B7 supertype-restricted Zaire ebolavirus T cell epitope canidates [42].

Agent based-model to simulate citrus-derived flavones as vaccine adjuvants against human papilloma virus 16; mouse in vivo result confirmation [43].

Investigation of differences in cellular A-to-I RNA editing activities upon infection with influenza A virus H1N1 and H3N2 [44].

RNA-Seq based analysis of differential innate immune response of RNA-Seq human cells infected with H1N1,H3N1, H5N1, HALo mutant and H7N9 and chicken and quail cells infected with H5N1 and H5N2 [45].

Cellular RNA editing analysis of C. albicans infected human epithelial cell lines and mouse in vivo infected tongue and kidney tissues [46].

Meta-genomics

16sPIP analysis pipeline for classification of 16S rDNA NGS data and screening of 346 clinically relevant pathogens [47].

CoMet binning workflow was used to assess contig coverage in combination with GC content for automated binning of a single and multi-strain metagenomic samples [48].

Metagenomic and -transcriptomic analysis of oolong teas to identify dominant microbial species and their anti-microbial peptides [49].

Computational pipeline ezTree infers marker genes and maximum likelihood phylogenetic trees from uncultivated prokaryotic genomes [50].

NGS genomics and transcript-omics

GTZ is a fast and lossless compression tool for cloud computing of FASTQ files; data transmission can overlap with compression; [51].

Coverage-dependent (from RNA sample concentration) RNA-Seq approach using a Bayesian method that infers the posterior distribution of a true gene count [52].

GT-WGS is a distributed whole-genome computing system based on Amazon Web Service cloud computing platform [53].

Pan-genome tool PGAP-X is a cross-platform software to analyze and visualize genome structure dynamics and and gene content [54].

Pará rubber (H. brasiliensis) genome transcriptome database [55].

Ontologies

InfAcrOnt tool can calculate similarities between terms across different ontologies and support the identifcation of novel relationships [56].

CroGO2 is an iterative ranking-based method that measures similarities of cross-categories GO terms using GO term level and interaction information in gene co-function networks [57].

Ontology of Chinese Medicine for Rheumatism represents 26 anti-rheumatism Chinese drugs together with their chemical ingredients, adverse effects and related information [58].

Post-translational modification sites

MDD-Carb is a tool for prediction of carbonylation sites utilizes maximal dependence decomposition and profile hidden Markov models [59].

SUCCESS is a SVM-based tool that predicts succinylation sites using structural and evolutionary information of amino acids [60].

CruxPTM is a platform for structure-based analysis post-translational modifications in context of PPI and drug binding [61].

Structural bio-informatics

Molecular dynamics analysis of charge states of balanol analogues that are ATP competitive inhibitors but nonselective for protein kinases A and C [62].

DeepSacon tool is a sparse autoencoder-based deep neural network to predict solvent accessibility and contact numbers [63].

R3D-BLAST2 is an improved search tool for similar RNA 3D substructures that can handle mmCIF files [64].