Applications of support vector machines to cancer classification with microarray data

Microarray gene expression data usually have a large number of dimensions, e.g., over ten thousand genes, and a small number of samples, e.g., a few tens of patients. In this paper, we use the support vector machine (SVM) for cancer classification with microarray data. Dimensionality reduction methods, such as principal components analysis (PCA), class-separability measure, Fisher ratio, and t-test, are used for gene selection. A voting scheme is then employed to do multi-group classification by k(k - 1) binary SVMs. We are able to obtain the same classification accuracy but with much fewer features compared to other published results.     

Ca handling alterations and vascular dysfunction in diabetes

More than 65% of patients with diabetes mellitus die from cardiovascular disease or stroke. Hyperglycemia, due to either reduced insulin secretion or reduced insulin sensitivity, is the hallmark feature of diabetes mellitus. Vascular dysfunction is a distinctive phenotype found in both types of diabetes and could be responsible for the high incidence of stroke, heart attack, and organ damage in diabetic patients. In addition to well-documented endothelial dysfunction, Ca²⁺ handling alterations in vascular smooth muscle cells (VSMCs) play a key role in the development and progression of vascular complications in diabetes. VSMCs provide not only structural integrity to the vessels but also control myogenic arterial tone and systemic blood pressure through global and local Ca²⁺ signaling. The Ca²⁺ signalosome of VSMCs is integrated by an extensive number of Ca²⁺ handling proteins (i.e. channels, pumps, exchangers) and related signal transduction components, whose function is modulated by endothelial effectors. This review summarizes recent findings concerning alterations in endothelium and VSMC Ca²⁺ signaling proteins that may contribute to the vascular dysfunction found in the diabetic condition.     

Genetic distances among the diabetic “Punjabis” and comparative populations from North India

Non-insulin dependent diabetes mellitus (NIDDM) is emerging as a major non-communicable disease in developing countries. Using principal component and Mahalanobis' generalized distance analyses, we examined gene frequency data at 3 polymorphic loci in sedentee urban controls, urban diabetics of the “Punjabi” population along with other comparative populations from North India. Mahalanobis generalized distance analysis suggests relative dissimilarity in the “Punjabi” disease subgroups indicating possible genetic heterogeneity. The genetic distance analysis also identified clinal variation along a north-west to south-east axis in North India.     

Diabetes Mellitus after Hematopoietic Stem Cell Transplantation

ObjectiveTo review the current literature on posttransplant diabetes mellitus after hematopoietic stem cell transplantation, including its epidemiologic features, transplant-related risk factors, and treatment.MethodsA literature search was conducted in PubMed for articles on diabetes mellitus after hematopoietic stem cell transplantation and effects of immunosuppressants on glucose metabolism.ResultsWithin 2 years after hematopoietic stem cell transplantation, up to 30% of patients may have diabetes. Although some of these cases resolve, the rates of diabetes and metabolic syndrome remain elevated in comparison with those in the nontransplant patient population during long-term follow-up. Traditional risk factors for diabetes as well as features related to the transplantation process, including immunosuppressive medications, are associated with posttransplant diabetes. Cardiovascular risk also appears to be increased in this population. Limited data are available on hypoglycemic agents for posttransplant diabetes; thus, treatment decisions must be based on safety, efficacy, and tolerability, with consideration of each patient’s transplant-related medications and comorbidities.ConclusionTreatment of diabetes mellitus in patients who have undergone hematopoietic stem cell transplantation necessitates attention to the posttransplant medication regimen and clinical course. Although no guidelines specific to treatment of posttransplant diabetes in this patient population currently exist, treatment to goals similar to those for nontransplant patients with diabetes should be considered in an attempt to help reduce long-term morbidity and mortality. (Endocr Pract. 2010;16:699-706)     

Computational methods for gene expression-based tumor classification

Gene expression profiles may offer more or additional information than classic morphologic- and histologic-based tumor classification systems. Because the number of tissue samples examined is usually much smaller than the number of genes examined, efficient data reduction and analysis methods are critical. In this report, we propose a principal component and discriminant analysis method of tumor classification using gene expression profile data. Expression of 2000 genes in 40 tumor and 22 normal colon tissue samples is used to examine the feasibility of gene expression-based tumor classification systems. Using this method, the percentage of correctly classified normal and tumor tissue was 87.0%. The combined approach using principal components and discriminant analysis provided superior sensitivity and specificity compared to an approach using simple differences in the expression levels of individual genes.     

Differential scanning calorimetry as a complementary diagnostic tool for the evaluation of biological samples

BackgroundDifferential scanning calorimetry (DSC) is a tool for measuring the thermal stability profiles of complex molecular interactions in biological fluids. DSC profiles (thermograms) of biofluids provide specific signatures which are being utilized as a new diagnostic approach for characterizing disease but the development of these approaches is still in its infancy.MethodsThis article evaluates several approaches for the analysis of thermograms which could increase the utility of DSC for clinical application. Thermograms were analyzed using localized thermogram features and principal components (PCs). The performance of these methods was evaluated alongside six models for the classification of a data set comprised of 300 systemic lupus erythematosus (SLE) patients and 300 control subjects obtained from the Lupus Family Registry and Repository (LFRR).ResultsClassification performance was substantially higher using the penalized algorithms relative to localized features/PCs alone. The models were grouped into two sets, the first having smoother solution vectors but lower classification accuracies than the second with seemingly noisier solution vectors.ConclusionsCoupling thermogram technology with modern classification algorithms provides a powerful diagnostic approach for analysis of biological samples. The solution vectors from the models may reflect important information from the thermogram profiles for discriminating between clinical groups.General significanceDSC thermograms show sensitivity to changes in the bulk plasma proteome that correlate with clinical status. To move this technology towards clinical application the development of new approaches is needed to extract discriminatory parameters from DSC profiles for the comparison and diagnostic classification of patients. This article is part of a Special Issue entitled Microcalorimetry in the BioSciences — Principles and Applications, edited by Fadi Bou-Abdallah.     

Anatomic changes in the biliary tract and pancreas in patients with diabetes mellitus diagnosed by endoscopic retrograde cholangiopancreatography]

Structural changes in both biliary tract and pancreas have been assessed with endoscopic retrograde cholangiopancreatography in 100 diabetic patients divided into subgroups depending on the type of diabetes mellitus, i.e. type I, type II and III-pancreatic. Control group included 100 randomly selected patients without diabetes mellitus in whom endoscopic retrograde cholangiopancreatography has been performed for various indications. Structural changes in the biliary tract and pancreas have been more frequent in diabetic patients than in the control group (47 and 75% vs 32 and 30%, respectively). Cholelithiasis has been noted in 27.8% of patients with type II diabetes mellitus and in 11.3% of patients with type I diabetes mellitus; obesity has been found in 57 and 12% of patients, respectively. Other biliary tract disorders, mainly in the form of segmental stenosis or dilatation of the common bile duct, have been more frequent in patients with type II diabetes mellitus. Pancreatic disorders, assessed with the aid of Cambridge classification, have been noted in all patients with pancreatic diabetes and in 80.7% of patients with diabetes mellitus type I. Incidence of so-called doubtful and mild disorders has been more frequent (22.2 and 24.1%, respectively) in patients with diabetes mellitus type II whereas "moderate" and "severe" disorders have been significantly less frequent (7.4 and 1.9% of patients). The results indicate, that endoscopic retrograde cholangiopancreatography is useful in the assessment of bile ducts structure and pancreatic exocrine activity in diabetic patients in whom disorders are more frequent.     

A fault identification and classification scheme for an automobile door assembly process

A process fault identification and classification scheme for an automobile door assembly process is presented based on multivariate in-line dimensional measurements and principal component factor analysis. First, the door assembly process and the dimensional measurement system are briefly introduced. Second, the technique of principal component factor analysis is presented for process fault identification. Process faults are summarized based on off-line identified case studies. Finally a machine classification scheme based on principal components and principal factors is presented and evaluated, using the pattern knowledge obtained off-line. This scheme is shown to be effective in classifying process faults using production data.     

Classification using functional data analysis for temporal gene expression data

MOTIVATION: Temporal gene expression profiles provide an important characterization of gene function, as biological systems are predominantly developmental and dynamic. We propose a method of classifying collections of temporal gene expression curves in which individual expression profiles are modeled as independent realizations of a stochastic process. The method uses a recently developed functional logistic regression tool based on functional principal components, aimed at classifying gene expression curves into known gene groups. The number of eigenfunctions in the classifier can be chosen by leave-one-out cross-validation with the aim of minimizing the classification error. RESULTS: We demonstrate that this methodology provides low-error-rate classification for both yeast cell-cycle gene expression profiles and Dictyostelium cell-type specific gene expression patterns. It also works well in simulations. We compare our functional principal components approach with a B-spline implementation of functional discriminant analysis for the yeast cell-cycle data and simulations. This indicates comparative advantages of our approach which uses fewer eigenfunctions/base functions. The proposed methodology is promising for the analysis of temporal gene expression data and beyond. AVAILABILITY: MATLAB programs are available upon request.     

A classification method based on principal components of SELDI spectra to diagnose of lung adenocarcinoma

Purpose

Lung cancer is the leading cause of cancer death worldwide, but techniques for effective early diagnosis are still lacking. Proteomics technology has been applied extensively to the study of the proteins involved in carcinogenesis. In this paper, a classification method was developed based on principal components of surface-enhanced laser desorption/ionization (SELDI) spectral data. This method was applied to SELDI spectral data from 71 lung adenocarcinoma patients and 24 healthy individuals. Unlike other peak-selection-based methods, this method takes each spectrum as a unity. The aim of this paper was to demonstrate that this unity-based classification method is more robust and powerful as a method of diagnosis than peak-selection-based methods.

Results

The results showed that this classification method, which is based on principal components, has outstanding performance with respect to distinguishing lung adenocarcinoma patients from normal individuals. Through leaving-one-out, 19-fold, 5-fold and 2-fold cross-validation studies, we found that this classification method based on principal components completely outperforms peak-selection-based methods, such as decision tree, classification and regression tree, support vector machine, and linear discriminant analysis.

Conclusions and Clinical Relevance

The classification method based on principal components of SELDI spectral data is a robust and powerful means of diagnosing lung adenocarcinoma. We assert that the high efficiency of this classification method renders it feasible for large-scale clinical use.     

The Taxonomy Statistic Uncovers Novel Clinical Patterns in a Population of Ischemic Stroke Patients

In this paper, we describe a simple taxonomic approach for clinical data mining elaborated by Marczewski and Steinhaus (M-S), whose performance equals the advanced statistical methodology known as the expectation-maximization (E-M) algorithm. We tested these two methods on a cohort of ischemic stroke patients. The comparison of both methods revealed strong agreement. Direct agreement between M-S and E-M classifications reached 83%, while Cohen’s coefficient of agreement was κ = 0.766(P < 0.0001). The statistical analysis conducted and the outcomes obtained in this paper revealed novel clinical patterns in ischemic stroke patients. The aim of the study was to evaluate the clinical usefulness of Marczewski-Steinhaus’ taxonomic approach as a tool for the detection of novel patterns of data in ischemic stroke patients and the prediction of disease outcome. In terms of the identification of fairly frequent types of stroke patients using their age, National Institutes of Health Stroke Scale (NIHSS), and diabetes mellitus (DM) status, when dealing with rough characteristics of patients, four particular types of patients are recognized, which cannot be identified by means of routine clinical methods. Following the obtained taxonomical outcomes, the strong correlation between the health status at moment of admission to emergency department (ED) and the subsequent recovery of patients is established. Moreover, popularization and simplification of the ideas of advanced mathematicians may provide an unconventional explorative platform for clinical problems.     

Small artery remodelling in diabetes

The aim of this article is to briefly review available data regarding changes in the structure of microvessels observed in patients with diabetes mellitus, and possible correction by effective treatment. The development of structural changes in the systemic vasculature is the end result of established hypertension. In essential hypertension, small arteries of smooth muscle cells are restructured around a smaller lumen and there is no net growth of the vascular wall, although in some secondary forms of hypertension, a hypertrophic remodelling may be detected. Moreover, in non-insulin-dependent diabetes mellitus a hypertrophic remodelling of subcutaneous small arteries is present. Indices of small resistance artery structure, such as the tunica media to internal lumen ratio, may have a strong prognostic significance in hypertensive and diabetic patients, over and above all other known cardiovascular risk factors. Therefore, regression of vascular alterations is an appealing goal of antihypertensive treatment. Different antihypertensive drugs seem to have different effect on vascular structure. In diabetic hypertensive patients, a significant regression of structural alterations of small resistance arteries with drugs blocking the renin–angiotensin system (angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers) was demonstrated. Alterations in the microcirculation represent a common pathological finding, and microangiopathy is one of the most important mechanisms involved in the development of organ damage as well as of clinical events in patients with diabetes mellitus. Renin–angiotensin system blockade seems to be effective in preventing/regressing alterations in microvascular structure.     

Incidence and manifestations of the heterozygosity of the gene causing diabetes]

A model of intermediate inheritance of diabetes mellitus is given, according to which juvenile diabetes developes in persons homozygous for the mutant gene, and a part of heterozygous carrier of the mutant gene has late diabetes mellitus. A large group of diabetes cases (about 50%), which are not due to direct diabetes heredity, is taken into account. Calculations are carried out on the basis of Hardy-Weinberg law from the patient questionaire data. The frequency of juvenile diabetes (developing within first thirty years) is 0,028% in the Leningrad population, the frequency of heterozygous diabetes gene carrier being 3,28%. Penetrancy of the diabetes gene in heterozygous state, as estimated from registrable (clinically expressed) forms of the disease, is 8,4%. The risk of juvenile diabetes in children, if one of the parents has juvenile diabetes, is 1,65% (3,3%-1/2), the frequency of heterozygous carriers in the population being 3,3%. The analysis of questionaire data of diabetes patients produced results, which were close to expected and indirectly confirmed the accurancy of the model proposed.     

Regularized common spatial patterns with subject-to-subject transfer of EEG signals

In the context of brain-computer interface (BCI) system, the common spatial patterns (CSP) method has been used to extract discriminative spatial filters for the classification of electroencephalogram (EEG) signals. However, the classification performance of CSP typically deteriorates when a few training samples are collected from a new BCI user. In this paper, we propose an approach that maintains or improves the recognition accuracy of the system with only a small size of training data set. The proposed approach is formulated by regularizing the classical CSP technique with the strategy of transfer learning. Specifically, we incorporate into the CSP analysis inter-subject information involving the same task, by minimizing the difference between the inter-subject features. Experimental results on two data sets from BCI competitions show that the proposed approach greatly improves the classification performance over that of the conventional CSP method; the transformed variant proved to be successful in almost every case, based on a small number of available training samples.     

Impact of Metabolic Syndrome on the Prognosis of Ischemic Stroke Secondary to Symptomatic Intracranial Atherosclerosis in Chinese Patients

Objectives

To analyze the effect of metabolic syndrome (MetS) on prognosis of ischemic stroke secondary to intracranial stenosis in Chinese patients.

Methods

A prospective cohort of 701 patients with ischemic stroke, caused by intracranial stenosis, were followed at 3-month intervals for 1 year to monitor development of recurrent stroke or death. Imaging was performed using magnetic resonance angiography. MetS was defined using International Diabetes Federation (IDF) criteria.

Results

MetS was identified in 26.0% of the cohort of stroke patients. Patients with MetS were more likely to be female, nonsmokers, and more likely to have a prior history of diabetes mellitus, high blood glucose and a family history of stroke than patients without MetS. During 1-year follow-up, patients with MetS had a non-significantly higher rate of stroke recurrence (7.1%) than patients without MetS (3.9%; P = 0.07). There was no difference in mortality (3.3% versus 3.5%, respectively). Multivariate Cox proportional hazards analysis (adjusting for gender, BMI, smoking, diabetes, and LDL-C) identified an association between that 1-year stroke recurrence and the presence of MetS (hazard ratio 2.30; 95% CI: 1.01–5.22) and large waist circumference (hazard ratio: 2.39; 95% CI: 1.05–5.42). However, multivariable analysis adjusting for the individual components of MetS found no significant associations between MetS and stroke recurrence. There were no associations between these parameters and mortality.

Conclusions

Chinese patients with symptomatic intracranial atherosclerosis who have MetS, are at higher risk of recurrent stroke than those without MetS. However, MetS was not predictive of stroke recurrence beyond its individual components and one-year mortality.     

Metabolic syndrome,Mediterranean diet,and polyphenols: Evidence and perspectives

Metabolic syndrome (MetS) is defined as the co-occurrence of metabolic risk factors that includes insulin resistance, hyperinsulinemia, impaired glucose tolerance, type 2 diabetes mellitus, dyslipidemia, and visceral obesity. The clinical significance of MetS consists of identifying a subgroup of patients sharing a common physiopathological state predisposing to chronic diseases. Clinical and scientific studies pinpoint lifestyle modification as an effective strategy aiming to reduce several features accountable for the risk of MetS onset. Among the healthy dietary patterns, the Mediterranean diet (MedDiet) emerges in terms of beneficial properties associated with longevity. Current evidence highlights the protective effect exerted by MedDiet on the different components of MetS. Interestingly, the effect exerted by polyphenols contained within the representative MedDiet components (i.e., olive oil, red wine, and nuts) seems to be accountable for the beneficial properties associated to this dietary pattern. In this review, we aim to summarize the principal evidence regarding the effectiveness of MedDiet–polyphenols in preventing or delaying the physiopathological components accountable for MetS onset. These findings may provide useful insights concerning the health properties of MedDiet–polyphenols as well as the novel targets destined to a tailored approach to MetS.     

广西地中海贫血112例临床分析

目的通过对广西112例地中海贫血病例的回顾性分析,了解地中海贫血患者的临床特点、治疗现状和并发症。方法用回顾性分析方法对广西112例地中海贫血患者(α中间型16例,β中间型30例,β重型66例)的临床特点、治疗现状和并发症进行分析。结果 (1)52例β重型地贫患者和所有需长时间频繁输血的中间型地贫患者输血前Hb90g/L。(2)不规范祛铁治疗的β中间型和β重型患者分别为84.0%和89.8%。(3)112例患者中,74例(67.9%)存在铁过载,2例α中间型,19例β中间型,53例β重型。(4)2例β中间型和1例β重型患者曾发生心力衰竭;2例β重型患者有继发性糖尿病;17例β中间型和46例β重型患者身材矮小,身材矮小与铁蛋白、输血前血红蛋白量相关(r=0.336,P0.001;r=-0.243,P=0.011)。结论本组β中间型地贫和β重型地贫患者输血和去铁治疗不规范,铁过载严重,身材矮小、心力衰竭等并发症发生率高。