O. Uğur Sezerman

Evolutionary selection of minimum number of features for classification of gene expression data using genetic algorithms

Selecting the most relevant factors from genetic profiles that can optimally characterize cellular states is of crucial importance in identifying complex disease genes and biomarkers for disease diagnosis and assessing drug efficiency. In this paper, we present an approach using a genetic algorithm for a feature subset selection problem that can be used in selecting the near optimum set of genes for classification of cancer data. In substantial improvement over existing methods, we classified cancer data with high accuracy with less features.

Testing robustness of relative complexity measure method constructing robust phylogenetic trees for Galanthus L. Using the relative complexity measure

BackgroundMost phylogeny analysis methods based on molecular sequences use multiple alignment where the quality of the alignment, which is dependent on the alignment parameters, determines the accuracy of the resulting trees. Different parameter combinations chosen for the multiple alignment may result in different phylogenies. A new non-alignment based approach, Relative Complexity Measure (RCM), has been introduced to tackle this problem and proven to work in fungi and mitochondrial DNA.ResultIn this work, we present an application of the RCM method to reconstruct robust phylogenetic trees using sequence data for genus Galanthus obtained from different regions in Turkey. Phylogenies have been analyzed using nuclear and chloroplast DNA sequences. Results showed that, the tree obtained from nuclear ribosomal RNA gene sequences was more robust, while the tree obtained from the chloroplast DNA showed a higher degree of variation.ConclusionsPhylogenies generated by Relative Complexity Measure were found to be robust and results of RCM were more reliable than the compared techniques. Particularly, to overcome MSA-based problems, RCM seems to be a reasonable way and a good alternative to MSA-based phylogenetic analysis. We believe our method will become a mainstream phylogeny construction method especially for the highly variable sequence families where the accuracy of the MSA heavily depends on the alignment parameters.

IDH-mutant glioma specific association of rs55705857 located at 8q24.21 involves MYC deregulation

The single nucleotide polymorphism rs55705857, located in a non-coding but evolutionarily conserved region at 8q24.21, is strongly associated with IDH-mutant glioma development and was suggested to be a causal variant. However, the molecular mechanism underlying this association has remained unknown. With a case control study in 285 gliomas, 316 healthy controls, 380 systemic cancers, 31 other CNS-tumors, and 120 IDH-mutant cartilaginous tumors, we identified that the association was specific to IDH-mutant gliomas. Odds-ratios were 9.25 (5.17–16.52; 95% CI) for IDH-mutated gliomas and 12.85 (5.94–27.83; 95% CI) for IDH-mutated, 1p/19q co-deleted gliomas. Decreasing strength with increasing anaplasia implied a modulatory effect. No somatic mutations were noted at this locus in 114 blood-tumor pairs, nor was there a copy number difference between risk-allele and only-ancestral allele carriers. CCDC26 RNA-expression was rare and not different between the two groups. There were only minor subtype-specific differences in common glioma driver genes. RNA sequencing and LC-MS/MS comparisons pointed to significantly altered MYC-signaling. Baseline enhancer activity of the conserved region specifically on the MYC promoter and its further positive modulation by the SNP risk-allele was shown in vitro. Our findings implicate MYC deregulation as the underlying cause of the observed association.

Biomarker discovery for toxicity

Toxicity biomarkers allow the safe evaluation of possible toxic effects of a substance in early phases of drug discovery. Finding the optimal subset of genes to use as biomarkers is an important problem. We tried evolutionary classification methods for finding biomarkers in hexachlorobenzene (HCB) toxicity using microarray data. We improve upon Kucukural et al. by modifying the algorithm to incrementally filter the features instead of generating new populations from scratch and by finding the common subset of features from multiple runs to be used as biomarkers. Using this modified genetic algorithm, we discovered gene sets of size 4 that were able to predict HCB exposure with >99% accuracy in 5-fold cross-validation tests. Repeating this process on independent test studies resulted in 14 biologically significant genes that predict exposure with 91% accuracy, surpassing other feature selection methods. Making use of these genes as biomarkers may allow us to detect hepatotoxic substances similar to HCB in a fast and cost-efficient manner when there are no emerging symptoms.

Discovery of Biomarkers for Hexachlorobenzene Toxicity Using Population Based Methods on Gene Expression Data

Discovering toxicity biomarkers is important in drug discovery to safely evaluate possible toxic effects of a substance in early phases. We tried evolutionary classification methods for selecting the important classifier genes in hexachlorobenzene toxicity using microarray data. Using modified genetic algorithms for selection of minimum number of features for classification of gene expression data, we discovered a number of gene sets of size 4 that were able to discriminate between the control and the hexachlorobenzene (HCB) exposed group of Brown-Norway rats with >99% accuracy in 5-fold cross-validation tests, whereas classification using all of the genes with SVM and other methods yielded results that vary between 48.48% to 81.81%. Making use of this small number of genes as biomarkers may allow us to detect toxicity of substances with mechanisms of toxicity similar to HCB in a fast and cost efficient manner when there are no emerging symptoms.

Discrimination of proteins using graph theoretic properties

Background Graph theoretic properties of proteins can be used to perceive the differences between correctly folded proteins and well designed decoy sets. Graphs are used to represent 3D protein structures. We used two different graph representations of protein structures which are Delaunay tessellations of proteins and contact map graphs. Graph theoretic properties for both graph types showed high classification accuracy to discrimination of proteins. Different type of linear classifiers and support vector classifier were used to classification of the protein structures. The best classifier accuracy was over 95% as shown in Table 1. The results showed that characteristic features of graph theoretic properties can be used many fields such as prediction of fold recognition, structure alignment and comparison, detection of similar domains and definition of structural motifs in high accuracy. Conclusion In this work we successfully showed that structural properties as well as potential scores can be used to discriminate native folds from the decoy sets. As far as graph types are concerned, the classification accuracy rates of the results obtained from contact map graphs are higher than the results obtained from Delaunay tessellated graphs for the same classification methods. Therefore contact map matrices are better representation method for protein structures. Support vector classifier and quadratic classifiers results are quite promising for the dataset which formed after outlier analysis. The accuracy rates are over 95%.

Predict permissive sites of protein for insertion domain

The procedure of domain insertion is proving to be an efficient way of creating modified proteins that can be used for multiple purposes. This procedure, however, requires certain calculations in order to determine permissive insertion sites of the proteins that will be modified. This paper aims to produce a scoring scheme based on the educated intuition of permissive sites to be flexible and accessible to predict the permissive sites of proteins for insertion domains.

Physico-chemical properties of DNA in phylogeny construction

Phylogenic analysis relies on alignment of related sequences from different species to obtain the distances between these species. The quality of the alignment and the distance measure would depend on the alignment parameters that are used. In this work, we propose to use Relative Complexity Measure (RCM) to find the distances between the sequences which is not a parameter dependent measure. We used DNA sequences from Candida species and phylogenetic trees were obtained using un-weighted pair-group with arithmetic mean method. We used three reduced alphabets for the DNA sequences which were clustered by taking into account different physicochemical properties of DNA. RCM gives as good results as the distance determination method and among the physicochemical properties, Keto/Amino grouping is found to give the most accurate tree which is topologically closest to the desired phylogeny.

Optimization of morphological data in numerical taxonomy analysis using genetic algorithms feature selection method

Studies in Numerical Taxonomy are carried out by measuring characters as much as possible. The workload over scientists and labor to perform measurements will increase proportionally with the number of variables (or characters) to be used in the study. However, some part of the data may be irrelevant or sometimes meaningless. Here in this study, we introduce an algorithm to obtain a subset of data with minimum characters that can represent original data. Morphological characters were used in optimization of data by Genetic Algorithms Feature Selection method. The analyses were performed on an 18 character*11 taxa data matrix with standardized continuous characters. The analyses resulted in a minimum set of 2 characters, which means the original tree based on the complete data can also be constructed by those two characters.