A multiple kernel support vector machine scheme for feature selection and rule extraction from gene expression data of cancer tissue

Advancing Computational Toxicology by Interpretable Machine Learning

Chemical toxicity evaluations for drugs, consumer products, and environmental chemicals have a critical impact on human health. Traditional animal models to evaluate chemical toxicity are expensive, time-consuming, and often fail to detect toxicants in humans. Computational toxicology is a promising alternative approach that utilizes machine learning (ML) and deep learning (DL) techniques to predict the toxicity potentials of chemicals. Although the applications of ML- and DL-based computational models in chemical toxicity predictions are attractive, many toxicity models are "black boxes" in nature and difficult to interpret by toxicologists, which hampers the chemical risk assessments using these models. The recent progress of interpretable ML (IML) in the computer science field meets this urgent need to unveil the underlying toxicity mechanisms and elucidate the domain knowledge of toxicity models. In this review, we focused on the applications of IML in computational toxicology, including toxicity feature data, model interpretation methods, use of knowledge base frameworks in IML development, and recent applications. The challenges and future directions of IML modeling in toxicology are also discussed. We hope this review can encourage efforts in developing interpretable models with new IML algorithms that can assist new chemical assessments by illustrating toxicity mechanisms in humans.

Dynamic structural analysis-based epitope prediction of Exendin-4 in aqueous solution

The study of epitopes has a broad range of applications in drug discovery, vaccine design, and immunotherapy. In this study, an epitope prediction method was developed based on the dynamic structure of protein antigens. Solvent accessible surface area, charge, and root mean square fluctuation were introduced as the key residue property parameters. The epitope prediction algorithm was established by constructing a three-parameter complex metrics of seven-peptide groups. The method was applied to predict the epitopes of Exendin-4, an effective antidiabetic drug. The epitopes of both the natural and C-terminal amidated forms of Exendin-4 were predicted and compared in their folded and intermediate states. In the folded state, the epitopes of natural Exendin-4 (His1-Phe6 and Asp9-Val19) were found to be nearly identical to the epitopes of C-terminal aminated Exendin-4 (His1-Thr7 and Asp9-Val19). In the intermediate state, however, the epitopes of natural Exendin-4 (His1-Gly4, Phe6 and Lys12-Arg20) covered fewer amino acids than the epitopes of C-terminal aminated Exendin-4 (His1-Gly4, Phe6, Asp9-Val19 and Trp25-Lys27). The comparison with the results from other prediction tools demonstrates the reliability of our predicted epitopes of Exendin-4.

A bibliometric and visual analysis of publications on artificial intelligence in colorectal cancer (2002-2022)

Background

Colorectal cancer (CRC) has the third-highest incidence and second-highest mortality rate of all cancers worldwide. Early diagnosis and screening of CRC have been the focus of research in this field. With the continuous development of artificial intelligence (AI) technology, AI has advantages in many aspects of CRC, such as adenoma screening, genetic testing, and prediction of tumor metastasis.

Objective

This study uses bibliometrics to analyze research in AI in CRC, summarize the field's history and current status of research, and predict future research directions.

Method

We searched the SCIE database for all literature on CRC and AI. The documents span the period 2002-2022. we used bibliometrics to analyze the data of these papers, such as authors, countries, institutions, and references. Co-authorship, co-citation, and co-occurrence analysis were the main methods of analysis. Citespace, VOSviewer, and SCImago Graphica were used to visualize the results.

Result

This study selected 1,531 articles on AI in CRC. China has published a maximum number of 580 such articles in this field. The U.S. had the most quality publications, boasting an average citation per article of 46.13. Mori Y and Ding K were the two authors with the highest number of articles. Scientific Reports, Cancers, and Frontiers in Oncology are this field's most widely published journals. Institutions from China occupy the top 9 positions among the most published institutions. We found that research on AI in this field mainly focuses on colonoscopy-assisted diagnosis, imaging histology, and pathology examination.

Conclusion

AI in CRC is currently in the development stage with good prospects. AI is currently widely used in colonoscopy, imageomics, and pathology. However, the scope of AI applications is still limited, and there is a lack of inter-institutional collaboration. The pervasiveness of AI technology is the main direction of future housing development in this field.

Human-in-the-loop machine learning: a state of the art

Researchers are defining new types of interactions between humans and machine learning algorithms generically called human-in-the-loop machine learning. Depending on who is in control of the learning process, we can identify: active learning, in which the system remains in control; interactive machine learning, in which there is a closer interaction between users and learning systems; and machine teaching, where human domain experts have control over the learning process. Aside from control, humans can also be involved in the learning process in other ways. In curriculum learning human domain experts try to impose some structure on the examples presented to improve the learning; in explainable AI the focus is on the ability of the model to explain to humans why a given solution was chosen. This collaboration between AI models and humans should not be limited only to the learning process; if we go further, we can see other terms that arise such as Usable and Useful AI. In this paper we review the state of the art of the techniques involved in the new forms of relationship between humans and ML algorithms. Our contribution is not merely listing the different approaches, but to provide definitions clarifying confusing, varied and sometimes contradictory terms; to elucidate and determine the boundaries between the different methods; and to correlate all the techniques searching for the connections and influences between them.

Artificial Intelligence Applied to Battery Research: Hype or Reality?

This is a critical review of artificial intelligence/machine learning (AI/ML) methods applied to battery research. It aims at providing a comprehensive, authoritative, and critical, yet easily understandable, review of general interest to the battery community. It addresses the concepts, approaches, tools, outcomes, and challenges of using AI/ML as an accelerator for the design and optimization of the next generation of batteries—a current hot topic. It intends to create both accessibility of these tools to the chemistry and electrochemical energy sciences communities and completeness in terms of the different battery R&D aspects covered.

Bearing remaining useful life prediction using support vector machine and hybrid degradation tracking model

Rolling element bearing is one of the critical components in rotating machines, and its running state determines machinery Remaining Useful Life (RUL). Estimating impending failure and predicting RUL of bearing is beneficial to schedule maintenance strategy and avoid abrupt shutdowns. This paper presents a novel method of RUL prediction of bearings, which can evaluate the degradation stage of bearings through dimensionless measurements and exploit the optimal RUL prediction through hybrid degradation tracing model in degradation stage. Two new measurements reflect the vibration intensity of bearings regarding normal vibration value. They can eliminate individual differences of bearings, improve sensitivity to the incipient defect of bearings, and reduce fluctuation. Moreover, they are helpful to detect the time to start prediction and set dimensionless failure threshold. SVM classifier is used to assess the degradation stage of bearing, which shows a high classification accuracy because of its excellent generalization ability and mathematical foundation. As input, the fitted measurements based on the generalized degradation model are used to train the SVM classifier. As output, five degradation stages are defined. However, actual measurements are used as inputs in the prediction process. According to the classification results, a hybrid degradation tracing model is utilized to exploit the optimal RUL prediction by tracking the degradation process of bearings. The proposed method is validated on the public IMS and PRONOSTIA bearing datasets, and its performance is compared with other methods on PRONOSTIA bearing datasets. The results show that the proposed approach is an effective way for RUL prediction of bearings within the prescribed error range. Given that the proposed measurements are dimensionless, this method can be applied under different operating conditions.

Artificial Intelligence Pipeline to Bridge the Gap between Bench Researchers and Clinical Researchers in Precision Medicine

Precision medicine informatics is a field of research that incorporates learning systems that generate new knowledge to improve individualized treatments using integrated data sets and models. Given the ever-increasing volumes of data that are relevant to patient care, artificial intelligence (AI) pipelines need to be a central component of such research to speed discovery. Applying AI methodology to complex multidisciplinary information retrieval can support efforts to discover bridging concepts within collaborating communities. This dovetails with precision medicine research, given the information rich multi-omic data that are used in precision medicine analysis pipelines. In this perspective article we define a prototype AI pipeline to facilitate discovering research connections between bioinformatics and clinical researchers. We propose building knowledge representations that are iteratively improved through AI and human-informed learning feedback loops supported through crowdsourcing. To illustrate this, we will explore the specific use case of nonalcoholic fatty liver disease, a growing health care problem. We will examine AI pipeline construction and utilization in relation to bench-to-bedside bridging concepts with interconnecting knowledge representations applicable to bioinformatics researchers and clinicians.

Refining wet lab experiments with in silico searches: A rational quest for diagnostic peptides in visceral leishmaniasis

Background

The search for diagnostic biomarkers has been profiting from a growing number of high quality sequenced genomes and freely available bioinformatic tools. These can be combined with wet lab experiments for a rational search. Improved, point-of-care diagnostic tests for visceral leishmaniasis (VL), early case detection and surveillance are required. Previous investigations demonstrated the potential of IgG1 as a biomarker for monitoring clinical status in rapid diagnostic tests (RDTs), although using a crude lysate antigen (CLA) as capturing antigen. Replacing the CLA by specific antigens would lead to more robust RDTs.

Methodology

Immunoblots revealed L. donovani protein bands detected by IgG1 from VL patients. Upon confident identification of these antigens by mass spectrometry (MS), we searched for evidence of constitutive protein expression and presence of antigenic domains or high accessibility to B-cells. Selected candidates had their linear epitopes mapped with in silico algorithms. Multiple high-scoring predicted epitopes from the shortlisted proteins were screened in peptide arrays. The most promising candidate was tested in RDT prototypes using VL and nonendemic healthy control (NEHC) patient sera.

Results

Over 90% of the proteins identified from the immunoblots did not satisfy the selection criteria and were excluded from the downstream epitope mapping. Screening of predicted epitope peptides from the shortlisted proteins identified the most reactive, for which the sensitivity for IgG1 was 84% (95% CI 60—97%) with Sudanese VL sera on RDT prototypes. None of the sera from NEHCs were positive.

Conclusion

We employed in silico searches to reduce drastically the output of wet lab experiments, focusing on promising candidates containing selected protein features. By predicting epitopes in silico we screened a large number of peptides using arrays, identifying the most promising one, for which IgG1 sensitivity and specificity, with limited sample size, supported this proof of concept strategy for diagnostics discovery, which can be applied to the development of more robust IgG1 RDTs for monitoring clinical status in VL.

Visceral leishmaniasis (VL) is a neglected tropical disease caused by protozoan parasites of the Leishmania donovani complex. Without treatment, VL is fatal. Although diagnostic techniques, mainly based on the detection of anti-Leishmania antibodies are available, invasive procedures such as microscopy from spleen or bone marrow aspirates are still required for the diagnosis of seronegative VL suspects, for the detection of recurrent cases and to confirm cure after successful treatment. Previous investigations showed the potential of IgG1 as a biomarker of post-chemotherapeutic relapse for VL in rapid diagnostic tests (RDTs) sensitised with crude lysate antigen (CLA). Here we employed in silico tools to search for desired protein features in a large number of L. donovani antigens detected by human IgG1 in western blots. We then employed prediction algorithms to profile epitopes from the shortlisted proteins. We screened a panel of high-scoring peptides in a high-throughput manner using arrays, with low reagent consumption. The most reactive peptide was adapted to RDTs, showing promising results of both sensitivity and specificity. This peptide has the potential of replacing the CLAs in IgG1 RDTs. Thus we believe that in silico tools can be used to optimise wet lab experiments for a rational search of biomarkers.

Statistical characterization and classification of colon microarray gene expression data using multiple machine learning paradigms

OBJECTIVE

A colon microarray data is a repository of thousands of gene expressions with different strengths for each cancer cell. It is necessary to detect which genes are responsible for cancer growth. This study presents an exhaustive comparative study of different machine learning (ML) systems which serves two major purposes: (a) identification of high risk differential genes using statistical tests and (b) development of a ML strategy for predicting cancer genes.

METHODS

Four statistical tests namely: Wilcoxon sign rank sum (WCSRS), t test, Kruskal-Wallis (KW), and F-test were adapted for cancerous gene identification using their p-values. The extracted gene set was used to classify cancer patients using ten classifiers namely: linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), naïve Bayes (NB), Gaussian process classification (GPC), support vector machine (SVM), artificial neural network (ANN), logistic regression (LR), decision tree (DT), Adaboost (AB), and random forest (RF). Performance was then evaluated using cross-validation protocols and standardized metrics viz. accuracy (ACC) and area under the curve (AUC).

RESULTS

The colon cancer dataset consists of 2000 genes from 62 patients (40 cancer vs. 22 control). The overall mean ACC of our ML system using all four statistical tests and all ten classifiers was 90.50%. The ML system showed an ACC of 99.81% using a combination WCSRS test and RF-based classifier. This is an improvement of 8% over previously published values in literature.

CONCLUSIONS

RF-based model with statistical tests for detection of high risk genes showed the best performance for accurate cancer classification in multi-center clinical trials.

Intelligent Techniques Using Molecular Data Analysis in Leukaemia: An Opportunity for Personalized Medicine Support System

The use of intelligent techniques in medicine has brought a ray of hope in terms of treating leukaemia patients. Personalized treatment uses patient's genetic profile to select a mode of treatment. This process makes use of molecular technology and machine learning, to determine the most suitable approach to treating a leukaemia patient. Until now, no reviews have been published from a computational perspective concerning the development of personalized medicine intelligent techniques for leukaemia patients using molecular data analysis. This review studies the published empirical research on personalized medicine in leukaemia and synthesizes findings across studies related to intelligence techniques in leukaemia, with specific attention to particular categories of these studies to help identify opportunities for further research into personalized medicine support systems in chronic myeloid leukaemia. A systematic search was carried out to identify studies using intelligence techniques in leukaemia and to categorize these studies based on leukaemia type and also the task, data source, and purpose of the studies. Most studies used molecular data analysis for personalized medicine, but future advancement for leukaemia patients requires molecular models that use advanced machine-learning methods to automate decision-making in treatment management to deliver supportive medical information to the patient in clinical practice.

A feature selection method based on multiple kernel learning with expression profiles of different types

Background

With the development of high-throughput technology, the researchers can acquire large number of expression data with different types from several public databases. Because most of these data have small number of samples and hundreds or thousands features, how to extract informative features from expression data effectively and robustly using feature selection technique is challenging and crucial. So far, a mass of many feature selection approaches have been proposed and applied to analyse expression data of different types. However, most of these methods only are limited to measure the performances on one single type of expression data by accuracy or error rate of classification.

Results

In this article, we propose a hybrid feature selection method based on Multiple Kernel Learning (MKL) and evaluate the performance on expression datasets of different types. Firstly, the relevance between features and classifying samples is measured by using the optimizing function of MKL. In this step, an iterative gradient descent process is used to perform the optimization both on the parameters of Support Vector Machine (SVM) and kernel confidence. Then, a set of relevant features is selected by sorting the optimizing function of each feature. Furthermore, we apply an embedded scheme of forward selection to detect the compact feature subsets from the relevant feature set.

Conclusions

We not only compare the classification accuracy with other methods, but also compare the stability, similarity and consistency of different algorithms. The proposed method has a satisfactory capability of feature selection for analysing expression datasets of different types using different performance measurements.

Electronic supplementary material

The online version of this article (doi:10.1186/s13040-017-0124-x) contains supplementary material, which is available to authorized users.

An Introduction to B-Cell Epitope Mapping and In Silico Epitope Prediction

Identification of B-cell epitopes is a fundamental step for development of epitope-based vaccines, therapeutic antibodies, and diagnostic tools. Epitope-based antibodies are currently the most promising class of biopharmaceuticals. In the last decade, in-depth in silico analysis and categorization of the experimentally identified epitopes stimulated development of algorithms for epitope prediction. Recently, various in silico tools are employed in attempts to predict B-cell epitopes based on sequence and/or structural data. The main objective of epitope identification is to replace an antigen in the immunization, antibody production, and serodiagnosis. The accurate identification of B-cell epitopes still presents major challenges for immunologists. Advances in B-cell epitope mapping and computational prediction have yielded molecular insights into the process of biorecognition and formation of antigen-antibody complex, which may help to localize B-cell epitopes more precisely. In this paper, we have comprehensively reviewed state-of-the-art experimental methods for B-cell epitope identification, existing databases for epitopes, and novel in silico resources and prediction tools available online. We have also elaborated new trends in the antibody-based epitope prediction. The aim of this review is to assist researchers in identification of B-cell epitopes.

Feature Selection Algorithm Based on Mutual Information and Lasso for Microarray Data

With the development of microarray technology, massive microarray data is produced by gene expression experiments, and it provides a new approach for the study of human disease. Due to the characteristics of high dimensionality, much noise and data redundancy for microarray data, it is difficult to my knowledge from microarray data profoundly and accurately,and it also brings enormous difficulty for information genes selection. Therefore, a new feature selection algorithm for high dimensional microarray data is proposed in this paper, which mainly involves two steps. In the first step, mutual information method is used to calculate all genes, and according to the mutual information value, information genes is selected as candidate genes subset and irrelevant genes are filtered. In the second step, an improved method based on Lasso is used to select information genes from candidate genes subset, which aims to remove the redundant genes. Experimental results show that the proposed algorithm can select fewer genes, and it has better classification ability, stable performance and strong generalization ability. It is an effective genes feature selection algorithm.

An automatic method for skeletal patterns classification using craniomaxillary variables on a Colombian population

BACKGROUND

The mandibular bone is an important part of the forensic facial reconstruction and it has the possibility of getting lost in skeletonized remains; for this reason, it is necessary to facilitate the identification process simulating the mandibular position only through craniomaxillary measures, for this task, different modeling techniques have been performed, but they only contemplate a straight facial profile that belong to skeletal pattern Class I, but the 24.5% corresponding to the Colombian skeletal patterns Class II and III are not taking into account, besides, craniofacial measures do not follow a parametric trend or a normal distribution.

OBJECTIVE

The aim of this study was to employ an automatic non-parametric method as the Support Vector Machines to classify skeletal patterns through craniomaxillary variables, in order to simulate the natural mandibular position on a contemporary Colombian sample.

MATERIALS AND METHODS

Lateral cephalograms (229) of Colombian young adults of both sexes were collected. Landmark coordinates protocols were used to create craniomaxillary variables. A Support Vector Machine with a linear kernel classifier model was trained on a subset of the available data and evaluated over the remaining samples. The weights of the model were used to select the 10 best variables for classification accuracy.

RESULTS

An accuracy of 74.51% was obtained, defined by Pr-A-N, N-Pr-A, A-N-Pr, A-Te-Pr, A-Pr-Rhi, Rhi-A-Pr, Pr-A-Te, Te-Pr-A, Zm-A-Pr and PNS-A-Pr angles. The Class Precision and the Class Recall showed a correct distinction of the Class II from the Class III and vice versa.

CONCLUSIONS

Support Vector Machines created an important model of classification of skeletal patterns using craniomaxillary variables that are not commonly used in the literature and could be applicable to the 24.5% of the contemporary Colombian sample.

Contribution of bioinformatics prediction in microRNA-based cancer therapeutics

Despite enormous efforts, cancer remains one of the most lethal diseases in the world. With the advancement of high throughput technologies massive amounts of cancer data can be accessed and analyzed. Bioinformatics provides a platform to assist biologists in developing minimally invasive biomarkers to detect cancer, and in designing effective personalized therapies to treat cancer patients. Still, the early diagnosis, prognosis, and treatment of cancer are an open challenge for the research community. MicroRNAs (miRNAs) are small non-coding RNAs that serve to regulate gene expression. The discovery of deregulated miRNAs in cancer cells and tissues has led many to investigate the use of miRNAs as potential biomarkers for early detection, and as a therapeutic agent to treat cancer. Here we describe advancements in computational approaches to predict miRNAs and their targets, and discuss the role of bioinformatics in studying miRNAs in the context of human cancer.

Application of genetic algorithms and constructive neural networks for the analysis of microarray cancer data

Background

Extracting relevant information from microarray data is a very complex task due to the characteristics of the data sets, as they comprise a large number of features while few samples are generally available. In this sense, feature selection is a very important aspect of the analysis helping in the tasks of identifying relevant genes and also for maximizing predictive information.

Methods

Due to its simplicity and speed, Stepwise Forward Selection (SFS) is a widely used feature selection technique. In this work, we carry a comparative study of SFS and Genetic Algorithms (GA) as general frameworks for the analysis of microarray data with the aim of identifying group of genes with high predictive capability and biological relevance. Six standard and machine learning-based techniques (Linear Discriminant Analysis (LDA), Support Vector Machines (SVM), Naive Bayes (NB), C-MANTEC Constructive Neural Network, K-Nearest Neighbors (kNN) and Multilayer perceptron (MLP)) are used within both frameworks using six free-public datasets for the task of predicting cancer outcome.

Results

Better cancer outcome prediction results were obtained using the GA framework noting that this approach, in comparison to the SFS one, leads to a larger selection set, uses a large number of comparison between genetic profiles and thus it is computationally more intensive. Also the GA framework permitted to obtain a set of genes that can be considered to be more biologically relevant. Regarding the different classifiers used standard feedforward neural networks (MLP), LDA and SVM lead to similar and best results, while C-MANTEC and k-NN followed closely but with a lower accuracy. Further, C-MANTEC, MLP and LDA permitted to obtain a more limited set of genes in comparison to SVM, NB and kNN, and in particular C-MANTEC resulted in the most robust classifier in terms of changes in the parameter settings.

Conclusions

This study shows that if prediction accuracy is the objective, the GA-based approach lead to better results respect to the SFS approach, independently of the classifier used. Regarding classifiers, even if C-MANTEC did not achieve the best overall results, the performance was competitive with a very robust behaviour in terms of the parameters of the algorithm, and thus it can be considered as a candidate technique for future studies.

PMBC: pattern mining from biological sequences with wildcard constraints

Patterns/subsequences frequently appearing in sequences provide essential knowledge for domain experts, such as molecular biologists, to discover rules or patterns hidden behind the data. Due to the inherent complex nature of the biological data, patterns rarely exactly reproduce and repeat themselves, but rather appear with a slightly different form in each of its appearances. A gap constraint (In this paper, a gap constraint (also referred to as a wildcard) is a character that can be substituted for any character predefined in an alphabet.) provides flexibility for users to capture useful patterns even if their appearances vary in the sequences. In order to find patterns, existing tools require users to explicitly specify gap constraints beforehand. In reality, it is often nontrivial or time-consuming for users to provide proper gap constraint values. In addition, a change made to the gap values may give completely different results, and require a separate time-consuming re-mining procedure. Therefore, it is desirable to automatically and efficiently find patterns without involving user-specified gap requirements. In this paper, we study the problem of frequent pattern mining without user-specified gap constraints and propose PMBC (namely P̲atternM̲ining from B̲iological sequences with wildcard C onstraints) to solve the problem. Given a sequence and a support threshold value (i.e. pattern frequency threshold), PMBC intends to discover all subsequences with their support values equal to or greater than the given threshold value. The frequent subsequences then form patterns later on. Two heuristic methods (one-way vs. two-way scans) are proposed to discover frequent subsequences and estimate their frequency in the sequences. Experimental results on both synthetic and real-world DNA sequences demonstrate the performance of both methods for frequent pattern mining and pattern frequency estimation.

CREDIT SCORING USING MULTI-KERNEL SUPPORT VECTOR MACHINE AND CHAOS PARTICLE SWARM OPTIMIZATION

Consumer credit scoring is considered as a crucial issue in the credit industry. SVM has been successfully utilized for classification in many areas including credit scoring. Kernel function is vital when applying SVM to classification problem for enhancing the prediction performance. Currently, most of kernel functions used in SVM are single kernel functions such as the radial basis function (RBF) which has been widely used. On the basis of the existing kernel functions, this paper proposes a multi-kernel function to improve the learning and generalization ability of SVM by integrating several single kernel functions. Chaos particle swarm optimization (CPSO) which is a kind of improved PSO algorithm is utilized to optimize parameters and to select features simultaneously. Two UCI credit data sets are used as the experimental data to evaluate the classification performance of the proposed method.

Generating balanced learning and test sets for function approximation problems

In function approximation problems, one of the most common ways to evaluate a learning algorithm consists in partitioning the original data set (input/output data) into two sets: learning, used for building models, and test, applied for genuine out-of-sample evaluation. When the partition into learning and test sets does not take into account the variability and geometry of the original data, it might lead to non-balanced and unrepresentative learning and test sets and, thus, to wrong conclusions in the accuracy of the learning algorithm. How the partitioning is made is therefore a key issue and becomes more important when the data set is small due to the need of reducing the pessimistic effects caused by the removal of instances from the original data set. Thus, in this work, we propose a deterministic data mining approach for a distribution of a data set (input/output data) into two representative and balanced sets of roughly equal size taking the variability of the data set into consideration with the purpose of allowing both a fair evaluation of learning's accuracy and to make reproducible machine learning experiments usually based on random distributions. The sets are generated using a combination of a clustering procedure, especially suited for function approximation problems, and a distribution algorithm which distributes the data set into two sets within each cluster based on a nearest-neighbor approach. In the experiments section, the performance of the proposed methodology is reported in a variety of situations through an ANOVA-based statistical study of the results.

Experiment specific expression patterns

The differential analysis of genes between microarrays from several experimental conditions or treatments routinely estimates which genes change significantly between groups. As genes are never regulated individually, observed behavior may be a consequence of changes in other genes. Existing approaches like co-expression analysis aim to resolve such patterns from a wide range of experiments. The knowledge of such a background set of experiments can be used to compute expected gene behavior based on known links. It is particularly interesting to detect previously unseen specific effects in other experiments. Here, a new method to spot genes deviating from expected behavior (PAttern DEviation SCOring--Padesco) is devised. It uses linear regression models learned from a background set to arrive at gene specific prediction accuracy distributions. For a given experiment, it is then decided whether each gene is predicted better or worse than expected. This provides a novel way to estimate the experiment specificity of each gene. We propose a validation procedure to estimate the detection of such specific candidates and show that these can be identified with an average accuracy of about 85%.

A glance at DNA microarray technology and applications

INTRODUCTION

Because of huge impacts of "OMICS" technologies in life sciences, many researchers aim to implement such high throughput approach to address cellular and/or molecular functions in response to any influential intervention in genomics, proteomics, or metabolomics levels. However, in many cases, use of such technologies often encounters some cybernetic difficulties in terms of knowledge extraction from a bunch of data using related softwares. In fact, there is little guidance upon data mining for novices. The main goal of this article is to provide a brief review on different steps of microarray data handling and mining for novices and at last to introduce different PC and/or web-based softwares that can be used in preprocessing and/or data mining of microarray data.

METHODS

To pursue such aim, recently published papers and microarray softwares were reviewed.

RESULTS

It was found that defining the true place of the genes in cell networks is the main phase in our understanding of programming and functioning of living cells. This can be obtained with global/selected gene expression profiling.

CONCLUSION

Studying the regulation patterns of genes in groups, using clustering and classification methods helps us understand different pathways in the cell, their functions, regulations and the way one component in the system affects the other one. These networks can act as starting points for data mining and hypothesis generation, helping us reverse engineer.

Lung cancer gene expression database analysis incorporating prior knowledge with support vector machine-based classification method

Background

A reliable and precise classification is essential for successful diagnosis and treatment of cancer. Gene expression microarrays have provided the high-throughput platform to discover genomic biomarkers for cancer diagnosis and prognosis. Rational use of the available bioinformation can not only effectively remove or suppress noise in gene chips, but also avoid one-sided results of separate experiment. However, only some studies have been aware of the importance of prior information in cancer classification.

Methods

Together with the application of support vector machine as the discriminant approach, we proposed one modified method that incorporated prior knowledge into cancer classification based on gene expression data to improve accuracy. A public well-known dataset, Malignant pleural mesothelioma and lung adenocarcinoma gene expression database, was used in this study. Prior knowledge is viewed here as a means of directing the classifier using known lung adenocarcinoma related genes. The procedures were performed by software R 2.80.

Results

The modified method performed better after incorporating prior knowledge. Accuracy of the modified method improved from 98.86% to 100% in training set and from 98.51% to 99.06% in test set. The standard deviations of the modified method decreased from 0.26% to 0 in training set and from 3.04% to 2.10% in test set.

Conclusion

The method that incorporates prior knowledge into discriminant analysis could effectively improve the capacity and reduce the impact of noise. This idea may have good future not only in practice but also in methodology.

Colon cancer prediction with genetic profiles using intelligent techniques

Micro array data provides information of expression levels of thousands of genes in a cell in a single experiment. Numerous efforts have been made to use gene expression profiles to improve precision of tumor classification. In our present study we have used the benchmark colon cancer data set for analysis. Feature selection is done using t-statistic. Comparative study of class prediction accuracy of 3 different classifiers viz., support vector machine (SVM), neural nets and logistic regression was performed using the top 10 genes ranked by the t-statistic. SVM turned out to be the best classifier for this dataset based on area under the receiver operating characteristic curve (AUC) and total accuracy. Logistic Regression ranks as the next best classifier followed by Multi Layer Perceptron (MLP). The top 10 genes selected by us for classification are all well documented for their variable expression in colon cancer. We conclude that SVM together with t-statistic based feature selection is an efficient and viable alternative to popular techniques.

Progressive changes in the development toward schizophrenia: studies in subjects at increased symptomatic risk

Although the underlying neurobiology of emerging psychotic disorders is not well understood, there is a growing conviction that the study of patients at clinical high risk for the illness will provide important insights. Further, a better understanding of the transition period may help the development of novel therapies. In this review, we summarize the extant neuroimaging and neuropsychological studies of people at clinical high risk for psychosis. By and large, there are few definitive markers that distinguish those who go on to develop the illness from those who do not. The 2 most consistently abnormal brain regions in schizophrenia research, the hippocampi and the lateral ventricles, are not significantly different from healthy controls prior to psychosis onset. However, frontal lobe measures (eg, cortical thickness in the anterior cingulate) do show promise, as do cognitive measures sensitive to prefrontal cortex dysfunction. Further, longitudinal magnetic resonance imaging findings in individuals at ultrahigh risk for developing a psychotic illness show that there are excessive neuroanatomical changes in those who convert to psychosis. These aberrant changes are observed most prominently in medial temporal and prefrontal cortical regions. While the pathological processes underlying such changes remain unclear, speculatively they may reflect anomalies in genetic and/or other endogenous mechanisms responsible for brain maturation, the adverse effects of intense or prolonged stress, or other environmental factors. Active changes during transition to illness may present the potential to intervene and ameliorate these changes with potential benefit clinically.