Current Bioinformatics - Volume 14, Issue 4, 2019

Background: To reduce the intensity of the work of doctors, pre-classification work needs to be issued. In this paper, a novel and related liver microscopic image classification analysis method is proposed. Objective: For quantitative analysis, segmentation is carried out to extract the quantitative information of special organisms in the image for further diagnosis, lesion localization, learning and treating anatomical abnormalities and computer-guided surgery. Methods: In the current work, entropy-based features of microscopic fibrosis mice’ liver images were analyzed using fuzzy c-cluster, k-means and watershed algorithms based on distance transformations and gradient. A morphological segmentation based on a local threshold was deployed to determine the fibrosis areas of images. Results: The segmented target region using the proposed method achieved high effective microscopy fibrosis images segmenting of mice liver in terms of the running time, dice ratio and precision. The image classification experiments were conducted using Gray Level Co-occurrence Matrix (GLCM). The best classification model derived from the established characteristics was GLCM which performed the highest accuracy of classification using a developed Support Vector Machine (SVM). The training model using 11 features was found to be accurate when only trained by 8 GLCMs. Conclusion: The research illustrated that the proposed method is a new feasible research approach for microscopy mice liver image segmentation and classification using intelligent image analysis techniques. It is also reported that the average computational time of the proposed approach was only 2.335 seconds, which outperformed other segmentation algorithms with 0.8125 dice ratio and 0.5253 precision.

Background: People in middle/later age often suffer from heart muscle damage due to coronary artery disease associated to myocardial infarction. In young people, the genetic forms of cardiomyopathies (heart muscle disease) are the utmost protuberant cause of myocardial disease. Objective: Accurate early detected information regarding the myocardial tissue structure is a key answer for tracking the progress of several myocardial diseases. Method: The present work proposes a new method for myocardium muscle texture classification based on entropy, homogeneity and on the texture unit-based texture spectrum approaches. Entropy and homogeneity are generated in moving windows of size 3x3 and 5x5 to enhance the texture features and to create the premise of differentiation of the myocardium structures. The texture is then statistically analyzed using the texture spectrum approach. Texture classification is achieved based on a fuzzy c–means descriptive classifier. The proposed method has been tested on a dataset of 80 echocardiographic ultrasound images in both short-axis and long-axis in apical two chamber view representations, for normal and infarct pathologies. Results: The noise sensitivity of the fuzzy c–means classifier was overcome by using the image features. The results established that the entropy-based features provided superior clustering results compared to homogeneity. Conclusion: Entropy image feature has a lower spread of the data in the clusters of healthy subjects and myocardial infarction. Also, the Euclidean distance function between the cluster centroids has higher values for both LAX and SAX views for entropy images.

Background: Stroke is one of the major causes for the momentary/permanent disability in the human community. Usually, stroke will originate in the brain section because of the neurological deficit and this kind of brain abnormality can be predicted by scrutinizing the periphery of brain region. Magnetic Resonance Image (MRI) is the extensively considered imaging procedure to record the interior sections of the brain to support visual inspection process. Objective: In the proposed work, a semi-automated examination procedure is proposed to inspect the province and the severity of the stroke lesion using the MRI. Method: Recently discovered heuristic approach called the Social Group Optimization (SGO) algorithm is considered to pre-process the test image based on a chosen image multi-thresholding procedure. Later, a chosen segmentation procedure is considered in the post-processing section to mine the stroke lesion from the pre-processed image. Results: In this paper, the pre-processing work is executed with the well known thresholding approaches, such as Shannon’s entropy, Kapur’s entropy and Otsu’s function. Similarly, the postprocessing task is executed using most successful procedures, such as level set, active contour and watershed algorithm. Conclusion: The proposed procedure is experimentally inspected using the benchmark brain stroke database known as Ischemic Stroke Lesion Segmentation (ISLES 2015) challenge database. The results of this experimental work authenticates that, Shannon’s approach along with the LS segmentation offers superior average values compared with the other approaches considered in this research work.

Background: The recently developed single-cell RNA sequencing (scRNA-seq) has attracted a great amount of attention due to its capability to interrogate expression of individual cells, which is superior to traditional bulk cell sequencing that can only measure mean gene expression of a population of cells. scRNA-seq has been successfully applied in finding new cell subtypes. New computational challenges exist in the analysis of scRNA-seq data. Objective: We provide an overview of the features of different similarity calculation and clustering methods, in order to facilitate users to select methods that are suitable for their scRNA-seq. We would also like to show that feature selection methods are important to improve clustering performance. Results: We first described similarity measurement methods, followed by reviewing some new clustering methods, as well as their algorithmic details. This analysis revealed several new questions, including how to automatically estimate the number of clustering categories, how to discover novel subpopulation, and how to search for new marker genes by using feature selection methods. Conclusion: Without prior knowledge about the number of cell types, clustering or semisupervised learning methods are important tools for exploratory analysis of scRNA-seq data.

Background: Chikungunya infection flare-ups have manifested in nations of Africa, Asia, and Europe including Indian and Pacific seas. It causes fever and different side effects include muscle torment, migraine, sickness, exhaustion and rash. It has turned into another, startling general medical issue in numerous tropical African and Asian countries and is presently being viewed as a genuine risk. No antiviral treatment or vaccine is yet available for this ailment. The current treatment is centered just on mitigating its side effects. Objective: The objective was to encourage the study on this viral pathogen, by the development of a database dedicated to Chikungunya Virus, that annotates and unifies the related data from various resources. Method: It undertook a consolidated approach for Chikungunya Virus genomic, proteomic, phylogenetics and therapeutic learning, involving the entire genome sequences and their annotation utilizing different in silico tools. Annotation included the information for CpG Island, usage bias, codon context and phylogenetic analysis at both the genome and proteome levels. Results: This database incorporates information of 41 strains of virus causing Chikungunya infection that can be accessed conveniently as well as downloaded effortlessly. Therapeutics section of this database contains data about B and T cell Epitopes, siRNAs and miRNAs that can be used as potential therapeutic targets. Conclusion: This database can be explored by specialists and established researchers around the world to assist their research on this non-treatable virus. It is a public database available from “www.chkv.in”.

Background: In recent years, more evidence have progressively indicated that Long non-coding RNAs (lncRNAs) play vital roles in wide-ranging human diseases, which can serve as potential biomarkers and drug targets. Comparing with vast lncRNAs being found, the relationships between lncRNAs and diseases remain largely unknown. Objective: The prediction of novel and potential associations between lncRNAs and diseases would contribute to dissect the complex mechanisms of disease pathogenesis. Method: In this paper, a new computational method based on Point Cut Set is proposed to predict LncRNA-Disease Associations (PCSLDA) based on known lncRNA-disease associations. Compared with the existing state-of-the-art methods, the major novelty of PCSLDA lies in the incorporation of distance difference matrix and point cut set to set the distance correlation coefficient of nodes in the lncRNA-disease interaction network. Hence, PCSLDA can be applied to forecast potential lncRNAdisease associations while known disease-lncRNA associations are required only. Results: Simulation results show that PCSLDA can significantly outperform previous state-of-the-art methods with reliable AUC of 0.8902 in the leave-one-out cross-validation and AUCs of 0.7634 and 0.8317 in 5-fold cross-validation and 10-fold cross-validation respectively. And additionally, 70% of top 10 predicted cancer-lncRNA associations can be confirmed. Conclusion: It is anticipated that our proposed model can be a great addition to the biomedical research field.

Background: Though after a start of genome sequencing most of the protein sequences are deposited in databases, some proteins remain to be unannotated and functionally uncharacterized. Chlamydia trachomatis L2C is a gram-negative pathogen bacterium involved in causing severe disorders like lymphogranuloma venereum, nongonococcal urethritis, and cervicitis. Objectives: Analyzing and annotating the hypothetical proteins can help to understand its pathogenicity and therapeutic hotspots. Its genome encodes a total of 221 hypothetical proteins and out of these, 14 hypothetical proteins are declared as virulent by virulence prediction server (VirulentPred). Methods: In this study, the functional and structural analysis was carried out by conserve domain finding servers, protein function annotators and physiochemical properties predictors. Proteinprotein interactions studies revealed the involvement of these virulent HPs in a number of pathways, which would be of interest for drug designers. Results: Classifier tool was used to classify the virulent hypothetical proteins into enzymes, membrane protein, transporter and regulatory protein groups. Conclusion: Our study would help to understand the mechanisms of pathogenesis and new potential therapeutic targets for a couple of diseases caused by C. trachomatis.

Background: Cancer subtype identification is an active research field which helps in the diagnosis of various cancers with proper treatments. Leukemia is one such cancer with various subtypes. High throughput technologies such as Deoxyribo Nucleic Acid (DNA) microarray are highly active in the field of cancer detection and classification alternatively. Objective: Yet, a precise analysis is important in microarray data applications as microarray experiments provide huge amount of data. Gene selection techniques promote microarray usage in the field of medicine. The objective of gene selection is to select a small subset of genes, which are the most informative in classification. Method: In this study, multi-objective evolutionary algorithm is used for gene subset selection in Leukemia classification. An initial redundant and irrelevant gene removal is followed by multiobjective evolutionary based gene subset selection. Gene subset selection highly influences the perfect classification. Thus, selecting the appropriate algorithm for subset selection is important. Results: The performance of the proposed method is compared against the standard genetic algorithm and evolutionary algorithm. Three Leukemia microarray datasets were used to evaluate the performance of the proposed method. Perfect classification was achieved for all the datasets only with few significant genes using the proposed approach. Conclusion: Thus, it is obvious that the proposed study perfectly classifies Leukemia with only few significant genes.

Background: Biomolecular-level event extraction is one of the most important branches of information extraction. With the rapid growth of biomedical literature, it is difficult for researchers to manually obtain information of interest, e.g. unknown information of threatening human disease or some biological processes. Therefore, researchers are interested in automatically acquiring information of biomolecular-level events. However, the annotated biomolecular-level event corpus is limited and highly imbalanced, which affects the performance of the classification algorithms and can even lead to over-fitting. Method: In this paper, a new approach using the Pairwise model and convolutional neural network for biomolecular-level event extraction is introduced. The method can identify more accurate positive instances from unlabeled data to enlarge the labeled data. First, unlabeled samples are categorized using the Pairwise model. Then, the shortest dependency path with additional information is generated. Furthermore, two input forms with a new representation of the convolutional neural network model, which are dependency word sequence and dependency relation sequence are presented. Finally, with the sample selection strategy, the expanded labeled samples from unlabeled domain corpus incrementally enlarge the training data to improve the performance of the classifier. Result & Conclusion: Our proposed method achieved better performance than other excellent systems. This is due to our new representation of generated short sentence and proposed sample selection strategy, which greatly improved the accuracy of classification. The extensive experimental results indicate that the new method can effectively inculcate unlabeled data to improve the performance of classifier for biomolecular-level events extraction.