Gene Expression Classification Research Articles

Weighted L1-norm Logistic Regression for Gene Selection of Microarray Gene Expression Classification

The classification of cancer is a significant application of the DNA microarray data. Gene selection methods are ordinarily used handle the issue of high-dimensionality of microarray data to enable experts to diagnose and classify cancer with high accuracy. The penalized logistic regression (PLR) technique is usually used in the dimensionality reduction of the high-dimensional gene expression data sets to remove irrelevant and redundant predictors from the binary logistic regression model. One of the regularization techniques used to achieve this goal is the least absolute shrinkage and selection operator (Lasso). However, this technique has been criticized for being biased in the selection of genes. The adaptive Lasso was usually proposed by assigning an initial weight to each gene to address the selection bias. This paper is concerned with adapting PLR to improve its capability in classification and gene selection, in the sense of accuracy, by introducing the one-dimensional weighted Mahalanobis distance (1-DWM) for each gene as an initial weight inside L1-norm. By experiments, this proposed method, denoted by adaptive penalized logistic regression (APLR), gives more accurate results compared with other famous methods in this regard. The proposed method is applied to some real high-dimensional gene expression data sets in order to demonstrate its efficiency in terms of classification accuracy and selection of gene. Therefore, the proposed method could be utilized in other studies implementing gene selection in the area of classification of high dimensional cancer data sets.

Autophagy-related gene expression classification defines three molecular subtypes with distinct clinical and microenvironment cell infiltration characteristics in colon cancer

BackgroundMultiple molecular subtypes with distinct clinical outcomes in colon cancer have been identified in recent years. Nonetheless, the autophagy-related molecular subtypes as well as its mediated tumor microenvironment (TME) cell infiltration characteristics have not been fully understood. MethodsBased on the seven colon cancer cohorts with 1580 samples, we performed a comprehensive genomic analysis to explore the molecular subtypes mediated by autophagy-related genes. The single-sample gene-set enrichment analysis (ssGSEA) was used to quantify the relative abundance of each cell infiltration in the TME. Unsupervised methods were used to perform autophagy subtype clustering. Least absolute shrinkage and selection operator regression (LASSO) was used to construct autophagy characterization score (APCS) signature. ResultsWe determined three distinct autophagy-related molecular subtypes in colon cancer. The three autophagy subtypes presented significant survival differences. Microenvironment analyses revealed the heterogeneous TME immune cell infiltration characterization between three subtypes. Cluster 1 autophagy subtype was characterized by abundant innate and adaptive immune cell infiltration. This subtype exhibited an enhanced stromal activity including activated pathways of epithelial-mesenchymal transition, TGF-β and angiogenesis, and an increased infiltration of fibroblasts and endothelial cells. The expression of immune checkpoint molecules was also significantly up-regulated, which may mediate immune escape in Cluster 1 subtype. Cluster 2 subtype was characterized by relatively lower TME immune cell infiltration and enhanced DNA damage repair pathways. Cluster 3 subtype was characterized by the suppression of immunity. Patients with high APCS, with poorer survival, presented a significantly positive correlation with TME stromal activity. Low APCS, relevant to activated damage repair pathways, showed enhanced responses to anti-PD-1/PD-L1 immunotherapy. Two immunotherapy cohorts confirmed patients with low APCS exhibited prominently enhanced clinical response and treatment advantages. ConclusionsThis study may help understand the molecular characterization of autophagy-related subtypes. We demonstrated the autophagy genes in colon cancer could drive the heterogeneity of TME immune cell infiltration. Our study represented a step toward personalized immunotherapy in colon cancer.

Combinatorial and statistical prediction of gene expression from haplotype sequence.

MotivationGenome-wide association studies (GWAS) have discovered thousands of significant genetic effects on disease phenotypes. By considering gene expression as the intermediary between genotype and disease phenotype, expression quantitative trait loci studies have interpreted many of these variants by their regulatory effects on gene expression. However, there remains a considerable gap between genotype-to-gene expression association and genotype-to-gene expression prediction. Accurate prediction of gene expression enables gene-based association studies to be performed post hoc for existing GWAS, reduces multiple testing burden, and can prioritize genes for subsequent experimental investigation.ResultsIn this work, we develop gene expression prediction methods that relax the independence and additivity assumptions between genetic markers. First, we consider gene expression prediction from a regression perspective and develop the HAPLEXR algorithm which combines haplotype clusterings with allelic dosages. Second, we introduce the new gene expression classification problem, which focuses on identifying expression groups rather than continuous measurements; we formalize the selection of an appropriate number of expression groups using the principle of maximum entropy. Third, we develop the HAPLEXD algorithm that models haplotype sharing with a modified suffix tree data structure and computes expression groups by spectral clustering. In both models, we penalize model complexity by prioritizing genetic clusters that indicate significant effects on expression. We compare HAPLEXR and HAPLEXD with three state-of-the-art expression prediction methods and two novel logistic regression approaches across five GTEx v8 tissues. HAPLEXD exhibits significantly higher classification accuracy overall; HAPLEXR shows higher prediction accuracy on approximately half of the genes tested and the largest number of best predicted genes () among all methods. We show that variant and haplotype features selected by HAPLEXR are smaller in size than competing methods (and thus more interpretable) and are significantly enriched in functional annotations related to gene regulation. These results demonstrate the importance of explicitly modeling non-dosage dependent and intragenic epistatic effects when predicting expression.Availability and implementationSource code and binaries are freely available at https://github.com/rapturous/HAPLEX.Supplementary information Supplementary data are available at Bioinformatics online.

Improvements in the Large p, Small n Classification Issue

Classifying gene expression data is known to contain keys for solving the fundamental problems in cancer studies. However, this issue is a complex task because of the large p, small n issue on gene expression data analysis. In this paper, we propose the improvements in the large p, small n classification issue for the study of human cancer. First, a new enhancing sample size method with generative adversarial network is proposed to improve classification algorithms. Second, we suggest a classification approach with over-sampling technique using features extracted by deep convolutional neural network. Numerical test results on fifty very-high-dimensional and low-sample-size gene expression data datasets from the Kent Ridge Biomedical and Array Expression repositories illustrate that the proposed models are more accurate than state-of-the-art classifying models. In addition, we also have explored the performance of support vector machines, k nearest neighbors and random forests, which have improved when apply our approaches.

DNA methylation profiling demonstrates superior diagnostic classification to RNA-sequencing in a case of metastatic meningioma

Meningiomas are the most common primary intracranial tumors, but meningioma metastases are rare. Accordingly, the clinical workup, diagnostic testing, and molecular classification of metastatic meningioma is incompletely understood. Here, we present a case report of multiply recurrent meningioma complicated by liver metastasis. We discuss the patient presentation, imaging findings, and conventional histopathologic characterization of both the intracranial lesion and the metastatic focus. Further, we perform multiplatform molecular profiling, comprised of DNA methylation arrays and RNA-sequencing, of six stereotactically-guided samples from the intracranial meningioma and a single ultrasound-guided liver metastasis biopsy. Our results show that DNA methylation clusters distinguish the liver metastasis from the intracranial meningioma samples, and identify a small focus of hepatocyte contamination with the liver biopsy. Nonetheless, DNA methylation-based classification accurately identifies the liver metastasis as a meningioma with high confidence. We also find that clustering of RNA-sequencing results distinguishes the liver metastasis from the intracranial meningiomas samples, but that differential gene expression classification is confounded by hepatocyte-specific gene expression programs in the liver metastasis. In sum, this case report sheds light on the comparative biology of intracranial and metastatic meningioma. Furthermore, our results support methylation-based classification as a robust method of diagnosing metastatic lesions, underscore the broad utility of DNA methylation array profiling in diagnostic pathology, and caution against the routine use of bulk RNA-sequencing for identifying tumor signatures in heterogeneous metastatic lesions.

Cancer molecular subtype classification from hypervolume-based discrete evolutionary optimization

High dimensionality and sample imbalance of gene expression data promote the development of effective algorithms for classifying gene expression data. To improve the ability to distinguish different subtypes of gene expression data, we devise a hypervolume-based discrete evolutionary optimization algorithm (HYBDEOA) in this paper. Four objectives, namely the number of genes, the accuracy, the relevance, and the redundancy, are optimized simultaneously to guide the evolution. Firstly, binary encoding is used to choose some features, projecting data onto different subspaces. After that, a discrete neighborhood operation is conducted to generate a new binary-mapped population. Combining the new population with the current population, we employ the hypervolume-based mechanism to select the Pareto solutions. Finally, a discrete mutation method is proposed to find promising solutions in the binary search space. To demonstrate the performance of HYBDEOA, we apply HYBDEOA to 55 synthetic datasets and 35 cancer gene expression datasets. Extensive experiments are also conducted to reveal the effectiveness and efficiency of HYBDEOA. The experimental results demonstrate that our proposed method is a parameter-less and robust algorithm, which can group gene expression data with a finer and more informative classification.

Rockburst assessment in deep geotechnical conditions using true-triaxial tests and data-driven approaches

Deep underground excavations in mining and civil engineering are subjected to high in-situ stresses which can cause rockburst. Rockburst is an instantaneous release of a large amount of strain energy stored in rockmass that can lead to injuries, deaths, and damage to infrastructures. Many studies have been done regarding rockburst, however, there is no practical model to predict the stress level that rockburst occurs (i.e. maximum rockburst stress) and its related risk (i.e. rockburst risk index) based on real rockburst tests, and the main rock mechanical properties. In this study, a comprehensive database of true-triaxial unloading tests on rocks having a wide range of properties was compiled. The agglomerative hierarchical clustering (AHC) analysis was carried out on the original database to evaluate the presence of natural groups and outliers. Then, the stepwise selection and elimination (SSE) procedure were employed for dimension reduction of the problem and identifying the most influential attributes on rockburst parameters. Afterward, two robust non-linear algorithms, including gene expression programming (GEP) and classification and regression tree (CART) were used to develop the predictive models for rockburst maximum stress and its risk index. The validation verification of the proposed models using several indices proved the high prediction performance of the developed non-linear models. Finally, a parametric analysis was carried out to evaluate the influence of each input parameter on the corresponding output. The proposed models in this study are practical and do not require any presupposition about rockburst mechanism, which makes them be used easily in practice by engineers at the design and progress stages of the underground projects.

Digital morphometry of tumor nuclei correlates to BAP-1 status, monosomy 3, gene expression class and survival in uveal melanoma

Cytologic features such as the shape and size of tumor cells can predict metastatic death in uveal melanoma and other cancers but suffer from poor reproducibility. In this study, we investigate the interobserver concordance of digital morphometry, and correlate the results with BRCA associated protein-1 (BAP-1) expression and BAP-1 gene mutation status, monosomy 3, gene expression classifications and patient survival in uveal melanoma. The average number of cells analyzed in each of 107 tumors, was 1957 (SD 349). Mean time consumption was less than 2.5 min per tumor. Identical morphometric classification was obtained for ≥85% of tumors in all twelve evaluated morphometric variables (κ 0.70–0.93). The mean nucleus area, nucleus perimeter, nucleus max caliper and nucleus to cell area ratio were significantly greater in tumors with low BAP-1 expression and gene expression class 2. Patients had significantly shorter survival if their tumors had low BAP-1 (Log-Rank p = 0.002), gene expression class 2 (p = 0.004), long nucleus perimeters (p = 0.031), long nucleus max calipers (p = 0.029) and high mean nucleus to cell area ratios (p = 0.041) as defined in a training cohort and then tested in a validation cohort. Long nucleus perimeters and long nucleus max calipers correlated with monosomy 3 (Pearson Chi-Square p = 0.006 and p = 0.009, respectively). Long nucleus perimeters also correlated with BAP-1 mutation (p = 0.017). We conclude that digital morphometry can be fast and highly reproducible, that for the first time, morphometry parameters can be objectively quantitated in thousands of cells at a time in sub-μm resolutions, and that variables describing the shape and size tumor nuclei correlate to BAP-1 status, monosomy 3, gene expression class as well as patient survival.

VisBicluster: A Matrix-Based Bicluster Visualization of Expression Data.

One of the main methods to analyze gene expression data is biclustering, a nonsupervised technique, which consists of selection subgroups of genes that co-expressed under subgroups of experimental conditions. A large number of biclustering algorithms have been developed to classify gene expression data. These algorithms can give as output a large number of overlapped biclusters, whose visualization still requires deeper studies. We present VisBicluster, a web-based interactive visualization tool for displaying biclustering results. The developed visualization technique consists of laying out the generated biclusters in a two-dimensional matrix where each bicluster is represented as a column and each overlap between a set of biclusters is represented as a row. A search interface for the user is developed to query the matrix of bicluster intersection and visualize the results matching the queries. Our tool supports many interactive features such as sorting, zooming, and details-on-demand. We proved the usefulness of VisBicluster with biclustering results from real and synthetic datasets. Besides, we performed a user study with 14 participants to illustrate the clarity and simplicity of overlap representation with our tool.

Discovering the Relationship between Heat-Stress Gene Expression and Gene SNPs Features using Rough Set Theory

Over the years of applying machine learning in bioinformatics, we have learned that scientists, working in many areas of life sciences, call for deeper knowledge of the modeled phenomenon than just the information used to classify the objects with a certain quality. As dynamic molecules of gene activities, transcriptome profiling by RNA sequencing (RNA-seq) is becoming increasingly popular, which not only measures gene expression but also structural variations such as mutations and fusion transcripts. Moreover, Single nucleotide polymorphisms (SNPs) are of great potential in genetics, breeding, ecological and evolutionary studies. Rough sets could be successfully employed to tackle various problems such as gene expression clustering and classification. This study provides general guidelines for accurate SNP discovery from RNA-seq data. Those SNPs annotations are used to find relation between their biological features and the differential expression of the genes to which those SNPs belong. Rough sets are utilized to define this kind of relationship into a finite set of rules. Set of (32) generated rules proved good results with strength, certainty and coverage evaluation terms. This strategy is applied to the analysis of SNPs in A. thaliana plant under heat-stress.

A Survey on Deep Learning Techniques for Prognosis and Diagnosis of Cancer from Microarray Gene Expression Data

The gene expression classification and identification from DNA microarray data is efficient technique for cancer diagnosis and prognosis for specific cancer subtypes. DNA microarray technology has great potential to discover information from expression levels of thousands of gene. The collection of significant genes which can improve the accuracy can give proper direction in early diagnosis of cancer. Cancer may be of different subtypes. Cancer detection from microarray gene expression data has major challenge of low sample size, high dimensionality and complexity of the data. There is a need for fast and computationally efficient method to deal with these kind of challenges. Deep Learning has succeeded in numerous fields such as image, video, speech, and text processing. Gene expression analysis is a unique challenge to Deep Learning for various cancer detection and prediction tasks in order to set specific biomarkers for different cancer subtypes. In this paper, we briefly discuss the strengths of different Deep Learning architectures for a cancer detection and prediction of various types of cancer through gene expression analysis.

Mining gene expression data using data mining techniques: A critical review

In the recent years, the emergence of data mining techniques in the research industries makes it possible to extract valuable information from huge volume of data. Good understanding of data mining techniques is necessary for research scholars and information industries to make use of this opportunity efficiently to improve the quality of their findings. Two most popular techniques of data mining are: clustering and classification. Clustering is an unsupervised learning method that groups the data based on distance and similarities among them. Classification assigns data items into target categories or classes with the aim of predicting the target class for each instance of data in a heterogeneous dataset accurately. This review focuses on current data mining techniques used in classification and clustering of gene expression dataset. Main attention is given to supervise and unsupervised methods that are being used recently to classify gene expression for the purpose of diagnosis and prognosis of terrifying diseases. In fact, gene expression data clustering offers a powerful approach to detect cancers from a given dataset. The various efficient algorithms available in the literature are analyzed to know their availability in the different situations.

Performance evaluation of naive Bayes and support vector machine in type 2 Diabetes Mellitus gene expression microarray data

Type 2 Diabetes Mellitus (T2DM) is a metabolic disorder that the number of diabetics increases every year. So that prevention is needed by knowing the trigger of T2DM. Gene expression microarray data contains information of gene that can be used to determine the causes of T2DM. It is necessary to use certain techniques to analyze gene expression microarray data because it has a large amount of data and attributes. This study aims to evaluate the performance of algorithms in classifying gene expression microarray data. Algorithms that were used in this study were Naive Bayes, and Support Vector Machine (SVM). SVM used many kernels function such as Linear, Radial Basis Function (RBF), Polynomial, and Sigmoid. Information gain was used to select the features in GSE18732 dataset by choosing top 10, 20, 30, 40, and 50 features. Performance of algorithms was evaluated and compared by using 30% testing set and 20% testing set. The results of the study indicated that SVM using Polynomial kernel had a high performance if it was compared to other algorithms. It achieved 98.15% accuracy using 30% testing set and achieved 100% accuracy using 20% testing set.

Multi-view based integrative analysis of gene expression data for identifying biomarkers

The widespread applications in microarray technology have produced the vast quantity of publicly available gene expression datasets. However, analysis of gene expression data using biostatistics and machine learning approaches is a challenging task due to (1) high noise; (2) small sample size with high dimensionality; (3) batch effects and (4) low reproducibility of significant biomarkers. These issues reveal the complexity of gene expression data, thus significantly obstructing microarray technology in clinical applications. The integrative analysis offers an opportunity to address these issues and provides a more comprehensive understanding of the biological systems, but current methods have several limitations. This work leverages state of the art machine learning development for multiple gene expression datasets integration, classification and identification of significant biomarkers. We design a novel integrative framework, MVIAm - Multi-View based Integrative Analysis of microarray data for identifying biomarkers. It applies multiple cross-platform normalization methods to aggregate multiple datasets into a multi-view dataset and utilizes a robust learning mechanism Multi-View Self-Paced Learning (MVSPL) for gene selection in cancer classification problems. We demonstrate the capabilities of MVIAm using simulated data and studies of breast cancer and lung cancer, it can be applied flexibly and is an effective tool for facing the four challenges of gene expression data analysis. Our proposed model makes microarray integrative analysis more systematic and expands its range of applications.

Novel hybrid DCNN–SVM model for classifying RNA-sequencing gene expression data

ABSTRACTIn recent years, cancer is one of the leading causes of death worldwide. Therefore, there are more and more studies that have been conducted to find effective solutions to diagnose and treat cancer. However, there are still many challenges in cancer treatment because possible causes of cancer are genetic disorders or epigenetic alterations in the cells. RNA sequencing is a powerful technique for gene expression profiling in model organisms and it is able to produce information for diagnosing cancer at the biomolecular level. Gene expression data are used to build a classification model which supports treatment of cancer. Nevertheless, its characteristic is very-high-dimensional data which lead to over-fitting issue of classifying model. In this paper, we propose a new gene expression classification model of support vector machines (SVM) using features extracted by deep convolutional neural network (DCNN). In our approach, the DCNN extracts latent features from gene expression data, then they are used in conjunction with SVM that efficiently classify RNA-Seq gene expression data. Numerical test results on RNA-Seq gene expression datasets from The Cancer Genome Atlas (TCGA) illustrate that our proposed algorithm is more accurate than state-of-the-art classifying models including DCNN, SVM and random forests.

The MMTV-Wnt1 murine model produces two phenotypically distinct subtypes of mammary tumors with unique therapeutic responses to an EGFR inhibitor.

ABSTRACTThe Wnt gene family encodes an evolutionarily conserved group of proteins that regulate cell growth, differentiation and stem cell self-renewal. Aberrant Wnt signaling in human breast tumors has been proposed as a driver of tumorigenesis, especially in the basal-like tumor subtype where canonical Wnt signaling is both enriched and predictive of poor clinical outcomes. The development of effective Wnt-based therapeutics, however, has been slowed in part by a limited understanding of the context-dependent nature with which these aberrations influence breast tumorigenesis. We previously reported that MMTV-Wnt1 mice, an established model for studying Wnt signaling in breast tumors, develop two subtypes of tumors by gene expression classification: Wnt1-EarlyEx and Wnt1-LateEx. Here, we extend this initial observation and show that Wnt1-EarlyEx tumors exhibit high expression of canonical Wnt, non-canonical Wnt, and EGFR signaling pathway signatures. Therapeutically, Wnt1-EarlyEx tumors showed a dynamic reduction in tumor volume when treated with an EGFR inhibitor. Wnt1-EarlyEx tumors had primarily Cd49fpos/Epcamneg FACS profiles, but it was not possible to serially transplant these tumors into wild-type FVB female mice. Conversely, Wnt1-LateEx tumors had a bloody gross pathology, which was highlighted by the presence of ‘blood lakes’ identified by H&E staining. These tumors had primarily Cd49fpos/Epcampos FACS profiles, but also contained a secondary Cd49fpos/Epcamneg subpopulation. Wnt1-LateEx tumors were enriched for activating Hras1 mutations and were capable of reproducing tumors when serially transplanted into wild-type FVB female mice. This study definitively shows that the MMTV-Wnt1 mouse model produces two phenotypically distinct subtypes of mammary tumors that differ in multiple biological aspects including sensitivity to an EGFR inhibitor.

Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders.

Genomics data provide great opportunities for translational research and the clinical practice, for example, for predicting disease stages. However, the classification of such data is a challenging task due to their high dimensionality, noise, and heterogeneity. In recent years, deep learning classifiers generated much interest, but due to their complexity, so far, little is known about the utility of this method for genomics. In this paper, we address this problem by studying a computational diagnostics task by classification of breast cancer and inflammatory bowel disease patients based on high‐dimensional gene expression data. We provide a comprehensive analysis of the classification performance of deep belief networks (DBNs) in dependence on its multiple model parameters and in comparison with support vector machines (SVMs). Furthermore, we investigate combined classifiers that integrate DBNs with SVMs. Such a classifier utilizes a DBN as representation learner forming the input for a SVM. Overall, our results provide guidelines for the complex usage of DBN for classifying gene expression data from complex diseases.

Random-Forest (RF) and Support Vector Machine (SVM) Implementation for Analysis of Gene Expression Data in Chronic Kidney Disease (CKD)

The application of mathematics in the field of bioinformatics has been widely developed. For example Support Vector Machines (SVM) and Random Forest (RF) are state of the art for classification of cancer in many applications. One of them is Chronic Kidney Disease (CKD). CKD is one of the kidney diseases that sufferers are increasing and have symptoms that are difficult to detect at first. Later, microarrays in gene expression are important tools for this approach. Microarrays gene expression provides an overview of all transcription activities in biological samples. The purpose of this research is a hybrid model combining Random Forest (RF) and Support Vector Machine (SVM) can be used to classify gene expression data. RF can highly accurate, generelize better and are interpretable and SVM (called RF-SVM) to effectively predict gene expression data with very high dimensions. In addition, from the simulation results on data from the Gene Expression Omnibus (GEO) database, it is shown that the proposed RF-SVM is a more accurate algorithm on CKD data than RFE-SVM.

Quantitative Temporal Proteomic Analysis of Vaccinia Virus Infection Reveals Regulation of Histone Deacetylases by an Interferon Antagonist

SummaryVaccinia virus (VACV) has numerous immune evasion strategies, including multiple mechanisms of inhibition of interferon regulatory factor 3 (IRF-3), nuclear factor κB (NF-κB), and type I interferon (IFN) signaling. Here, we use highly multiplexed proteomics to quantify ∼9,000 cellular proteins and ∼80% of viral proteins at seven time points throughout VACV infection. A total of 265 cellular proteins are downregulated >2-fold by VACV, including putative natural killer cell ligands and IFN-stimulated genes. Two-thirds of these viral targets, including class II histone deacetylase 5 (HDAC5), are degraded proteolytically during infection. In follow-up analysis, we demonstrate that HDAC5 restricts replication of both VACV and herpes simplex virus type 1. By generating a protein-based temporal classification of VACV gene expression, we identify protein C6, a multifunctional IFN antagonist, as being necessary and sufficient for proteasomal degradation of HDAC5. Our approach thus identifies both a host antiviral factor and a viral mechanism of innate immune evasion.

A new distributed feature selection technique for classifying gene expression data

Classification of gene expression data is a pivotal research area that plays a substantial role in diagnosis and prediction of diseases. Generally, feature selection is one of the extensively used techniques in data mining approaches, especially in classification. Gene expression data are usually composed of dozens of samples characterized by thousands of genes. This increases the dimensionality coupled with the existence of irrelevant and redundant features. Accordingly, the selection of informative genes (features) becomes difficult, which badly affects the gene classification accuracy. In this paper, we consider the feature selection for classifying gene expression microarray datasets. The goal is to detect the most possibly cancer-related genes in a distributed manner, which helps in effectively classifying the samples. Initially, the available huge amount of considered features are subdivided and distributed among several processors. Then, a new filter selection method based on a fuzzy inference system is applied to each subset of the dataset. Finally, all the resulted features are ranked, then a wrapper-based selection method is applied. Experimental results showed that our proposed feature selection technique performs better than other techniques since it produces lower time latency and improves classification performance.