Microarray Gene Expression Datasets Research Articles

A novel immune prognostic index for stratification of high-risk patients with early breast cancer

The prognostic value of current multigene assays for breast cancer is limited to hormone receptor-positive, human epidermal growth factor receptor 2-negative early breast cancer. Despite the prognostic significance of immune response-related genes in breast cancer, immune gene signatures have not been incorporated into most multigene assays. Here, using public gene expression microarray datasets, we classified breast cancer patients into three risk groups according to clinical risk and proliferation risk. We then developed the immune prognostic index based on expression of five immune response-related genes (TRAT1, IL2RB, CTLA4, IGHM and IL21R) and lymph node status to predict the risk of recurrence in the clinical and proliferation high-risk (CPH) group. The 10-year probability of disease-free survival (DFS) or distant metastasis-free survival (DMFS) of patients classified as high risk according to the immune prognostic index was significantly lower than those of patients classified as intermediate or low risk. Multivariate analysis revealed that the index is an independent prognostic factor for DFS or DMFS. Moreover, the C-index revealed that it is superior to clinicopathological variables for predicting prognosis. Its prognostic significance was also validated in independent datasets. The immune prognostic index identified low-risk patients among patients classified as CPH, regardless of the molecular subtype of breast cancer, and may overcome the limitations of current multigene assays.

A systems biology approach for identifying key genes and pathways of gastric cancer using microarray data

Gastric cancer (GC) is a serious malignant disorder with a mortality rate of approximately 10%. It is therefore necessary to identify prognostic biomarkers and drug targets to improve the clinical outcome of patients with GC. The present research aims to distinguish the key genes and the relevant pathways using a microarray gene expression dataset that facilitates the understanding of molecular mechanisms. The identified genes may be used as the therapeutic biomarker of GC. Seven important pathways, like, ECM-receptor interaction, Gastric acid secretion, PI3K-Akt signaling pathway, Focal adhesion, Amoebiasis, Gastric acid secretion, and Collecting duct acid secretion, linked with gastric cancer are identified. The 11 genes associated with these pathways are considered as the potential targets for gastric cancer drug discovery. Finally, survival analysis based on the expression of these genes generates 7 key genes (COL1A1, COL1A2, THBS2, FN1, SPP1, ATP4A, and ATP4B) having poor overall survival. We have also constructed a protein-protein interaction (PPI) network of the differentially expressed genes (DEGs) using STRING and 57 hub genes are identified using Cytoscape. Survival analysis of these hub genes yields 16 genes that are highly correlated with the poor survival of GC patients. The current study would help understand the disease mechanisms and accelerate its diagnosis.

WITHDRAWN: Bacterial foraging optimization algorithm based feature selection for microarray data classification

An effective method is needed to extract knowledge and useful information from microarray gene expression datasets. Datasets of microarray gene expression typically consist of large numbers of genes and less samples, making it very difficult to extract data and trends present in the data. To choose the genes with maximum variations that are the most informative and important genes, gene selection is therefore carried out. The genes whose value in the samples does not differ much and remain roughly the same are excluded and are not used for the classifier design. Recent research has focused on the use of data mining to classify microarray data and to identify genes that are differentially regulated by different diseases. In this paper, gene selection was based on the technique of bacterial foraging optimization (BFO). E.coli bacteria develop at a very rapid rate due to the suitable conditions and adequate food, founded on the drinking performance of the Bacteria E.coli. The bacteria are moving into nutrient areas very rapidly and seeking to escape harmful substances. The bacterial movements are called taxes.

Overexpression of GDP dissociation inhibitor 1 gene associates with the invasiveness and poor outcomes of colorectal cancer

ABSTRACT GDP dissociation inhibitor (GDI) regulates the GDP/GTP exchange reaction of most Rab proteins by inhibiting GDP dissociation. This study evaluated the potential prognostic and predictive value of GDI1 in colorectal cancer (CRC). To address the prognostic power of GDI1, we performed individual and pooled survival analyses on six independent CRC microarray gene expression datasets. GDI1-enriched signatures were also analyzed. Kaplan–Meier and Cox proportional analyses were employed for survival analysis. An immunohistochemistry (IHC) analysis was performed to validate the clinical relevance and prognostic significance of the GDI1 protein level in CRC tissue samples. The results revealed that GDI1 mRNA level was significantly linked with the aggressiveness of CRC, which is compatible with gene set enrichment analysis. A meta-analysis and pooled analysis demonstrated that a higher mRNA GDI1 expression was dramatically correlated with a worse survival in a dose-dependent manner in CRC patients. Further IHC analysis validated that the protein expression of GDI1 in both cytoplasm and membrane also significantly impacted the outcome of CRC patients. In CRC patients with stage III, chemotherapy significantly reduced the relative risk of death in low-GDI1 subgroup (hazard ratio (HR) = 0.22; 95% confidence interval (95% CI) 0.09–0.56, p = 0.0003), but not in high-GDI1 subgroup (HR = 0.63; 95% CI 0.35–1.14, p = 0.1137). Therefore, both high mRNA and protein levels of GDI1 were significantly related to poor outcomes in CRC patients. GD11 may serve as a prognostic biomarker for CRC.

A hybrid gene selection model for molecular breast cancer classification using a deep neural network

Microarray-based gene expression outlining portrays a dominant part in a healthier understanding of breast cancer. From the large quantum of data, a powerful technique is required to understand and extract the required information. The molecular subtype extraction is one of such important information regarding breast cancer, which is very crucial in defining its treatment strategy. This manuscript has formulated a deep neural network-based model for molecular classification of breast cancer. The proposed model exploits pre-processing steps along with the hybrid approach of filter and wrapper-based feature selection to extract relevant genes. The extracted genes are evaluated using various machine learning approaches where it is observed that selected features are successful in solving this multiclass problem. Using the proposed hybrid model, we have achieved the highest accuracy with six microarray datasets. The model outperforms magnificently in standings of sensitivity, f-measure, specificity, MCC and recall. Hence, deep neural network is identified as the best efficient classifiers concluding brilliant performance with all the selected microarray gene expression datasets for a range of selected genes.

Prognostic significance of KRT19 in Lung Squamous Cancer.

Backgroud: Keratin 19 (KRT19) is the intermediate filament that constitutes the cytoskeleton and regulates cell-cycle and cell death.Objective: We aimed to assess whether KRT19 was involved in lung cancer development.Methods: The expression of KRT19 in lung cancer was evaluated from mRNA expression on open databse and protein abundance on tumor tissue array.Results: Using open microarray gene expression datasets and differential expression analysis, we found that KRT19 was upregulated in lung cancer compared with normal tissue. Further analysis suggested that KRT19 mRNA expression was correlated with tumor progression and overall survival in lung cancer patients. As KRT19 was overexpressed in adenocarcinoma (AC) and squamous cell carcinoma (SCC), we examined the prognostic value of KRT19 protein abundance by tissue microarray (TMA). The results suggested that protein expression of KRT19 was significantly associated with overall survival of SCC.Conclusions: Giving the prognostic role of KRT19 in lung cancer, KRT19 could be considered as an potential molecular marker in lung cancer, especially in SCC.

A comprehensive investigation on liver regeneration: a meta-analysis and systems biology approach.

Aim of the studyLiver regeneration is one of the essential fields of regenerative medicine as a branch of tissue engineering and molecular biology that draws global researchers’ attention. This study aims to conduct a systematic review and meta-analysis on the high-throughput gene expression microarray dataset of liver regeneration on the NCBI-GEO database to identify the significant genes and signaling pathways and confirm the genes from literature studies on associated diseases.Material and methodsWe thoroughly searched the NCBI-GEO database to retrieve and screen the GEO microarray datasets’ contents. Due to the inclusion of different species in eligible GEO datasets in the meta-analysis, the list of significant genes for the random-effects model were identified. Moreover, we carried out detailed gene analyses for three main gene ontology components and the KEGG signaling pathway. Furthermore, we investigated the possibility of genes’ association with liver cancer through the Kaplan-Meier plot.ResultsThe random-effects model from six eligible GEO datasets identified 71 genes with eight down-regulated and 63 up-regulated genes. The target genes are involved in various cellular functions such as cell proliferation, cell death, and cell cycle control. Finally, we noted that 58 out of 71 genes are associated with different types of diseases related explicitly to other liver and inflammation diseases.ConclusionsThe current study assessed various GEO datasets at the early stages of liver regeneration with promising results. The present systematic review and meta-analysis results are beneficial for future novel drug design and discovery specifically for patients in the liver transplantation process.

WITHDRAWN: A novel wrapper FS based on binary swallow swarm optimization with score-based criteria fusion for gene expression microarray data

In recent days, Gene Expression Microarray (GEM) data increases its significance to diagnose the diseases appropriately. The key issue of analyzing gene expression information for huge amount of genes with less samples it to extract the disease-related data from the large number of unwanted information and noise. Feature choice, removing unwanted and inappropriate genes was a major process for investigating this challenge. So, the Feature Selection (FS) and categorization are the primary tasks in analyzing the GEM. In this article, a Swallow Swarm Optimization (SSO) with Score-based Criteria Fusion (Optimized SCF (OSCF)) wrapper FStechnique is developed to predict the cancer with a high classification efficiency. Novel wrapper FStechnique follows the procedure of Binary Weight SSO (BWSSO) and the fitness is evaluated by combining the SCF and classification accuracy. OSCF algorithm fuses SCF and classification accuracy to measure the genes significance and redundancy for verifying the genes dependency. It could improve the efficiency of the classifiers such as K-Nearest Neighbours (KNN), Support Vector Machine (SVM) and Recursive Neural Networks (RNN), and evaluate its improvement using four different GEM datasets (Prostate cancer, Small Round Blue Cell Tumours (SRBCT), Leukemia, and Lymphoma). Experimental results verify that the OSCF technique shows better results when compared to conventional techniques based on precision, recall, accuracy and Area Under Curve (AUC).

Comparative analysis of transcriptomic profile, histology, and IDH mutation for classification of gliomas

Gliomas are currently classified through integration of histology and mutation information, with new developments in DNA methylation classification. However, discrepancies exist amongst the major classification methods. This study sought to compare transcriptome-based classification to the established methods. RNAseq and microarray data were obtained for 1032 gliomas from the TCGA and 395 gliomas from REMBRANDT. Data were analyzed using unsupervised and supervised learning and other statistical methods. Global transcriptomic profiles defined four transcriptomic glioma subgroups with 91.4% concordance with the WHO-defined mutation subtypes. Using these subgroups, 168 genes were selected for the development of 1000 linear support vector classifiers (LSVC). Based on plurality voting of 1000 LSVC, the final ensemble classifier confidently classified all but 17 TCGA gliomas to one of the four transcriptomic profile (TP) groups. The classifier was validated using a gene expression microarray dataset. TP1 cases include IDHwt, glioblastoma high immune infiltration and cellular proliferation and poor survival prognosis. TP2a is characterized as IDHmut-codel, oligodendrogliomas with high tumor purity. TP2b tissue is mostly composed of neurons and few infiltrating malignant cells. TP3 exhibit increased NOTCH signaling, are astrocytoma and IDHmut-non-codel. TP groups are highly concordant with both WHO integrated histology and mutation classification as well as methylation-based classification of gliomas. Transcriptomic profiling provides a robust and objective method to classify gliomas with high agreement to the current WHO guidelines and may provide additional survival prediction to the current methods.

Early Diagnosis of Pancreatic Ductal Adenocarcinoma by Combining Relative Expression Orderings With Machine-Learning Method.

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive and lethal cancer deeply affecting human health. Diagnosing early-stage PDAC is the key point to PDAC patients’ survival. However, the biomarkers for diagnosing early PDAC are inexact in most cases. Therefore, it is highly desirable to identify an effective PDAC diagnostic biomarker. In the current work, we designed a novel computational approach based on within-sample relative expression orderings (REOs). A feature selection technique called minimum redundancy maximum relevance was used to pick out optimal REOs. We then compared the performances of different classification algorithms for discriminating PDAC and its adjacent normal tissues from non−PDAC tissues. The support vector machine algorithm is the best one for identifying early PDAC diagnostic biomarker. At first, a signature composed of nine gene pairs was acquired from microarray gene expression data sets. These gene pairs could produce satisfactory classification accuracy up to 97.53% in fivefold cross-validation. Subsequently, two types of data from diverse platforms, namely, microarray and RNA-Seq, were used to validate this signature. For microarray data, all (100.00%) of 115 PDAC tissues and all (100.00%) of 31 PDAC adjacent normal tissues were correctly recognized as PDAC. In addition, 88.24% of 17 non-PDAC (normal or pancreatitis) tissues were correctly classified. For the RNA-Seq data, all (100.00%) of 177 PDAC tissues and all (100.00%) of 4 PDAC adjacent normal tissues were correctly recognized as PDAC. Validation results demonstrated that the signature had a good cross-platform effect for early detection of PDAC. This work developed a new robust signature that might be a promising biomarker for early PDAC diagnosis.

Diagnostic biomarker candidates for pulpitis revealed by bioinformatics analysis of merged microarray gene expression datasets

BackgroundPulpitis is an inflammatory disease, the grade of which is classified according to the level of inflammation. Traditional methods of evaluating the status of dental pulp tissue in clinical practice have limitations. The rapid and accurate diagnosis of pulpitis is essential for determining the appropriate treatment. By integrating different datasets from the Gene Expression Omnibus (GEO) database, we analysed a merged expression matrix of pulpitis, aiming to identify biological pathways and diagnostic biomarkers of pulpitis.MethodsBy integrating two datasets (GSE77459 and GSE92681) in the GEO database using the sva and limma packages of R, differentially expressed genes (DEGs) of pulpitis were identified. Then, the DEGs were analysed to identify biological pathways of dental pulp inflammation with Gene Ontology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and Gene Set Enrichment Analysis (GSEA). Protein–protein interaction (PPI) networks and modules were constructed to identify hub genes with the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) and Cytoscape.ResultsA total of 470 DEGs comprising 394 upregulated and 76 downregulated genes were found in pulpitis tissue. GO analysis revealed that the DEGs were enriched in biological processes related to inflammation, and the enriched pathways in the KEGG pathway analysis were cytokine-cytokine receptor interaction, chemokine signalling pathway and NF-κB signalling pathway. The GSEA results provided further functional annotations, including complement system, IL6/JAK/STAT3 signalling pathway and inflammatory response pathways. According to the degrees of nodes in the PPI network, 10 hub genes were identified, and 8 diagnostic biomarker candidates were screened: PTPRC, CD86, CCL2, IL6, TLR8, MMP9, CXCL8 and ICAM1.ConclusionsWith bioinformatics analysis of merged datasets, biomarker candidates of pulpitis were screened and the findings may be as reference to develop a new method of pulpitis diagnosis.

Statistical Analysis of Microarray Data Clustering using NMF, Spectral Clustering, Kmeans, and GMM.

In unsupervised learning literature, the study of clustering using microarray gene expression datasets has been extensively conducted with nonnegative matrix factorization (NMF), spectral clustering, kmeans, and gaussian mixture model (GMM)are some of the most used methods. However, there is still a limited number of works that utilize statistical analysis to measure the significances of performance differences between these methods. In this paper, statistical analysis of performance differences between ten NMF, six spectral clustering, four GMM, and the standard kmeans algorithms in clustering eleven publicly available microarray gene expression datasets with the number of clusters ranges from two to ten is presented. The experimental results show that statistically NMFs and kmeans have similar performances and outperform spectral clustering. As spectral clustering can be used to uncover hidden manifold structures, the underperformance of spectral methods leads us to question whether the datasets have manifold structures. Visual inspection using multidimensional scaling plots indicates that such structures do not exist. Moreover, as the plots indicate that clusters in some datasets have elliptical boundaries, GMM methods are also utilized. The experimental results show that GMM methods outperform the other methods to some degree, and thus imply that the datasets follow gaussian distributions.

Abstract 4309: A novel immune prognostic index for the stratification of high-risk patients with early breast cancer

Abstract The prognostic or predictive value of current commercial multigene assays for breast cancer are limited to hormone receptor-positive, human epidermal growth factor receptor 2-negative (HR+/HER2-) early breast cancer. Despite the prognostic or predictive significance of immune response-related genes in breast cancer, immune gene signatures have not been incorporated into most current assays. Here, we developed a novel immune gene based-prognostic model to predict the risk of recurrence in high-risk women with early breast cancer. Using public gene expression microarray datasets for breast cancer, we classified breast cancer patients into three risk groups according to clinical risk and proliferation risk, and identified the most significant five immune response-related genes (TRAT1, IL2RB, CTLA4, IGHM and IL21R) associated with clinical outcome in the clinical and proliferation high-risk (CPH) subgroup. Immune prognostic index based on the combination of the expression of five immune response-related genes and lymph node status was developed to predict the risk of recurrence in high-risk breast cancer patients and its prognostic significance was validated in independent validation datasets. Disease-free survival (DFS) or distant metastasis-free survival (DMFS) of patients classified as high risk according to the immune prognostic index was significantly lower than those of patients classified as intermediate and low risk in the CPH group. Multivariate analysis demonstrated that the immune prognostic index is an independent prognostic factor for DFS or DMFS. Moreover, c-index showed that the immune prognostic index is superior to clinicopathological variables in predicting the prognosis. Our immune index was also prognostic in independent validation datasets. Our immune prognostic index identified low-risk patients among patients classified as CPH group in all molecular subtypes of breast cancer and may be used to overcome the limitations of the current multigene assay. Citation Format: Hannah Lee, Mi Jeong Kwon, Beom-Mo Koo, Jinil Han, Young Kee Shin. A novel immune prognostic index for the stratification of high-risk patients with early breast cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4309.

Classification of gene expression patterns using a novel type-2 fuzzy multigranulation-based SVM model for the recognition of cancer mediating biomarkers

In this article, we propose a novel type-2 fuzzy multigranulation-based SVM model for gene expression pattern classification on human breast cancer dataset. Firstly, a type-2 fuzzy multigranulation system has been designed for the classification task dealing with noisy and nonlinear microarray gene expression patterns. Thereafter, the fuzzy if–then rules have been devised on the feature vectors of microarray to enable accurate bilinear classification process. The fuzzy if–then rules in the domain of type-2 fuzzy multigranulation system are able to identify efficient expression patterns that have been deferentially expressed from normal state to carcinogenic state. The proposed method reduces the structural complexity of the fuzzy if–then rules (type 1) since it works on upper and lower membership functions instead of a single membership function. In addition, a fuzzy rough approximation has been utilized in the model to reduce the computational cost. Lastly, the association among genes consisting of significantly different expression patterns from normal state to malignant state has been recognized with respect to their nature. The effectiveness of the proposed method has been implemented on eight microarray gene expression datasets for human breast cancer patients. Moreover, we have validated the results by F-score and NCBI database which signify that the proposed model performs better in comparison with the existing methods.

Identification of the aberrantly methylated differentially expressed genes in proliferative diabetic retinopathy

Diabetic retinopathy (DR) is the most common complication of diabetes. Proliferative DR (PDR) is a more advanced stage of DR, which can cause severe impaired vision and even blindness. However, the precise pathological mechanisms of PDR remain unknown. DNA methylation serves an important role in the initiation and progression of numerous types of disease including PDR. The purpose of this study was to identify the aberrantly methylated differentially expressed genes (DEGs) as potential therapeutic targets of PDR. The gene expression microarray dataset GSE60436 and the methylation profiling microarray dataset GSE57362 were used to determine the aberrantly methylated DEGs in PDR, utilizing normal retinas as controls and fibrovascular membranes (FVMs) in patients with PDR as PDR samples. The functional term and signaling pathway enrichment analysis of the selected genes were subsequently performed. In addition, protein-protein interaction (PPI) networks were constructed to determine the hub genes, and the network of transcriptional factor (TF) and target hub genes was also analyzed. In total, 132 hypomethylated genes were found to be upregulated, whereas 172 hypermethylated genes were discovered to be downregulated in PDR. The hypomethylated upregulated genes were found to be enriched in the pathways, such as “cell-substrate adhesion”, “adherens junction”, “cell adhesion molecule binding” and “extracellular matrix receptor interactions”. Meanwhile, the hypermethylated downregulated genes were enriched in the pathways, such as “visual perception”, “presynapse” and the “synaptic vesicle cycle”. Based on the PPI analysis, a total of eight hub genes were identified: CTGF, SERPINH1, LOX, RBP3, OTX2, RPE65, OPN1SW and NRL. It was hypothesized that the aberrant methylation of these genes might be related to the possible pathophysiology of PDR. An important transcriptional factor, TFDP1, was discovered to share the closest interactions with the hub genes from the gene-TF network. In conclusion, the present study identified an association among DNA methylation and gene expression in PDR using bioinformatics analysis, and identified the hub genes which might be potential methylation-based diagnosis and treatment targets for PDR in the near future.

Identification of a five-gene signature for predicting the progression and prognosis of stage I endometrial carcinoma.

Uterine corpus endometrial carcinoma (UCEC) is often diagnosed at an early clinical stage based on abnormal vaginal bleeding. However, the prognosis of UCEC is poor. The present study was conducted to identify novel tumor grade-related genes with the potential to predict the prognosis and progression of UCEC. A total of three gene expression microarray datasets were downloaded from the Gene Expression Omnibus database, and one RNA-sequencing dataset with corresponding clinical information of patients with UCEC was obtained from The Cancer Genome Atlas database. In summary, 1,447 differentially expressed genes (DEGs) were identified between endometrial cancerous tissues and normal endometrial tissues. Weighted gene co-expression network analysis was performed to assess the associations between DEGs and clinical traits. In total, five genes were found to be highly associated with the tumorigenesis and prognosis of UCEC. Among them, BUB1 mitotic checkpoint serine/threonine kinase B, cyclin B1, cell-division cycle protein 20 and non-SMC condensing I complex subunit G were involved in cell cycle regulation pathways, and DLG-associated protein 5 was involved in the Notch receptor 3 signaling pathway based on functional enrichment analyses. Of the five genes, four were highly expressed in endometrial cancerous tissues compared with normal endometrial tissues at the protein level. In addition, the higher expression of these genes predicted a higher tumor grade and worse overall survival. In conclusion, the present study revealed a 5-gene signature that can be used to predict the progression of UCEC.

A new hybrid stability measure for feature selection

Feature Selection (FS) algorithms are applied in bioinformatics applications to identify the disease causing genes. Performance of such algorithms is measured in terms of accuracy of the model and stability of FS algorithms. Stability evaluates the identical replication of feature sets obtained after every execution. Recently research has shown that a stability measure must satisfy set of properties like, fully defined, monotonicity, boundedness, deterministic maximum stability, and correction for chance. Among the existing stability measures, only Nogueira’s frequency based stability measure satisfies all the required properties. However, frequency based stability measures fail to discriminate among the cases when overall frequency of features are same. In order to address this issue, the paper proposes a hybrid similarity based stability measure which satisfies all the desirable properties, as mentioned earlier. The proposed stability measure is unique as it is the first similarity based stability measure that satisfies all the required properties. Also, all these essential properties are mathematically established. Further, the paper also proposes a combination of frequency based and similarity based measure which preserves all the aspects of both the approaches. The work presented also analyzes the stability performance of LASSO and Elastic Net, using synthetic and microarray gene expression datasets. Elastic Net depicts higher stability and selection of relevant features.

Effect of the Abnormal Expression of BMP-4 in the Blood of Diabetic Patients on the Osteogenic Differentiation Potential of Alveolar BMSCs and the Rescue Effect of Metformin: A Bioinformatics-Based Study.

The success rate of oral implants is lower in type 2 diabetes mellitus (T2DM) patients than in nondiabetic subjects; functional impairment of bone marrow-derived mesenchymal stem cells (BMSCs) is an important underlying cause. Many factors in the blood can act on BMSCs to regulate their biological functions and influence implant osseointegration, but which factors play important negative roles in T2DM patients is still unclear. This study is aimed at screening differentially expressed genes in the blood from T2DM and nondiabetic patients, identifying which genes impact the osteogenic differentiation potential of alveolar BMSCs in T2DM patients, exploring drug intervention regimens, and providing a basis for improving implant osseointegration. Thus, a whole-blood gene expression microarray dataset (GSE26168) of T2DM patients and nondiabetic controls was analyzed. Based on Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) results, differentially expressed genes and signaling pathways related to BMSC osteogenic differentiation were screened, and major risk genes were extracted based on the mean decrease Gini coefficient calculated using the random forest method. Bone morphogenetic protein-4 (BMP-4), with significantly low expression in T2DM blood, was identified as the most significant factor affecting BMSC osteogenic differentiation potential. Subsequently, metformin, a first-line clinical drug for T2DM treatment, was found to improve the osteogenic differentiation potential of BMSCs from T2DM patients via the BMP-4/Smad/Runx2 signaling pathway. These results demonstrate that low BMP-4 expression in the blood of T2DM patients significantly hinders the osteogenic function of BMSCs and that metformin is effective in counteracting the negative impact of BMP-4 deficiency.

MISSING DATA ESTIMATION WITH GAM (GENERALIZED ADDITIVE MODEL) REGRESSION AND FUZZY C-MEANS DIMENSIONALITY REDUCTION FOR MICROARRAY GENE EXPRESSION DATA

Missing data estimation is one of the important task performed during the pre-processing of data before applying the data mining algorithms. Generally the data which is used for data mining or machine learning consisting of complete data without missing entries. But usually the data used for data mining or machine learning occurred with missing values due to some technical or general reasons. Performance of data mining or machine learning entirely depends on the form of data and its existing values. If the dataset have huge missing entries then it drastically affect the performance of the algorithms and degrades the accuracy. So it is important to estimate and impute these missing values before computation. Several data imputation algorithms were proposed in the literature. In this paper, we proposed a new methodology by reducing the dimensionality of the data by using fuzzy-C means clustering and later used the generalized additive model to estimate the missing values in the data set. Three real time microarray gene expression datasets were used in this paper and performance of these are compared with existing algorithms.

Predicting autism spectrum disorder from associative genetic markers of phenotypic groups using machine learning

Machine learning is a discipline of artificial intelligence, geared towards the development of various critical applications. Due to its high precision, it is widely adopted in the process of extracting useful hidden patterns and valuable insights from complex data structures. Data extracted from the real-time environment might contain some irrelevant information. The presence of noise in the data degrades the model performance. Gene expression is an important source, carries the genetic information of species. Gene expression pattern reveals the significant relationship between genes associated with several diseases. But due to irregular molecular interactions and reactions occurs during the transcription process, the gene expressions are minimally affected. It causes a detrimental effect on the identification of biological markers of the diseases. To address this problem, a novel gene selection strategy is proposed to identify the candidate gene biomarkers from the genomic data. Signal to Noise ratio with logistic sigmoid function, Hilbert–Schmidt Independence Criterion Lasso, and regularized genetic algorithm amalgamation finds the optimal features. The proposed system is tested with the microarray gene expression dataset of autism spectrum disorder (ASD), accessed from gene expression omnibus repository. FAM104B, CCNDBP1, H1F0, ZER1 are identified as the candidate biomarkers of ASD. The methodical performance evaluation of the proposed model is examined with widely used machine learning algorithms. The proposed methodology enhanced the prediction rate of ASD and attained an accuracy of 97.62%, outperformed existing methods. Also, this system could act as a significant tool to assist the medical practitioners for accurate ASD diagnosis.