In silico multiple-targets identification for heme detoxification in the human malaria parasite Plasmodium falciparum.

Detoxification of hemoglobin byproducts or free heme is an essential step and considered potential targets for anti-malaria drug development. However, most of anti-malaria drugs are no longer effective due to the emergence and spread of the drug resistant malaria parasites. Therefore, it is an urgent need to identify potential new targets and even for target combinations for effective malaria drug design. In this work, we reconstructed the metabolic networks of Plasmodium falciparum and human red blood cells for the simulation of steady mass and flux flows of the parasite's metabolites under the blood environment by flux balance analysis (FBA). The integrated model, namely iPF-RBC-713, was then adjusted into two stage-specific metabolic models, which first was for the pathological stage metabolic model of the parasite when invaded the red blood cell without any treatment and second was for the treatment stage of the parasite when a drug acted by inhibiting the hemozoin formation and caused high production rate of heme toxicity. The process of identifying target combinations consisted of two main steps. Firstly, the optimal fluxes of reactions in both the pathological and treatment stages were computed and compared to determine the change of fluxes. Corresponding enzymes of the reactions with zero fluxes in the treatment stage but non-zero fluxes in the pathological stage were predicted as a preliminary list of potential targets in inhibiting heme detoxification. Secondly, the combinations of all possible targets listed in the first step were examined to search for the best promising target combinations resulting in more effective inhibition of the detoxification to kill the malaria parasites. Finally, twenty-three enzymes were identified as a preliminary list of candidate targets which mostly were in pyruvate metabolism and citrate cycle. The optimal set of multiple targets for blocking the detoxification was a set of heme ligase, adenosine transporter, myo-inositol 1-phosphate synthase, ferrodoxim reductase-like protein and guanine transporter. In conclusion, the method has shown an effective and efficient way to identify target combinations which are obviously useful in the development of novel antimalarial drug combinations.

A high performance prediction of HPV genotypes by Chaos game representation and singular value decomposition.

BACKGROUND: Human Papillomavirus (HPV) genotyping is an important approach to fight cervical cancer due to the relevant information regarding risk stratification for diagnosis and the better understanding of the relationship of HPV with carcinogenesis. This paper proposed two new feature extraction techniques, i.e. ChaosCentroid and ChaosFrequency, for predicting HPV genotypes associated with the cancer. The additional diversified 12 HPV genotypes, i.e. types 6, 11, 16, 18, 31, 33, 35, 45, 52, 53, 58, and 66, were studied in this paper. In our proposed techniques, a partitioned Chaos Game Representation (CGR) is deployed to represent HPV genomes. ChaosCentroid captures the structure of sequences in terms of centroid of each sub-region with Euclidean distances among the centroids and the center of CGR as the relations of all sub-regions. ChaosFrequency extracts the statistical distribution of mono-, di-, or higher order nucleotides along HPV genomes and forms a matrix of frequency of dots in each sub-region. For performance evaluation, four different types of classifiers, i.e. Multi-layer Perceptron, Radial Basis Function, K-Nearest Neighbor, and Fuzzy K-Nearest Neighbor Techniques were deployed, and our best results from each classifier were compared with the NCBI genotyping tool. RESULTS: The experimental results obtained by four different classifiers are in the same trend. ChaosCentroid gave considerably higher performance than ChaosFrequency when the input length is one but it was moderately lower than ChaosFrequency when the input length is two. Both proposed techniques yielded almost or exactly the best performance when the input length is more than three. But there is no significance between our proposed techniques and the comparative alignment method. CONCLUSIONS: Our proposed alignment-free and scale-independent method can successfully transform HPV genomes with 7,000 - 10,000 base pairs into features of 1 - 11 dimensions. This signifies that our ChaosCentroid and ChaosFrequency can be served as the effective feature extraction techniques for predicting the HPV genotypes.

New feature selection for gene expression classification based on degree of class overlap in principal dimensions.

Micro-array data are typically characterized by high dimensional features with a small number of samples. Several problems in identifying genes causing diseases from micro-array data can be transformed into the problem of classifying the features extracted from gene expression in micro-array data. However, too many features can cause low prediction accuracy as well as high computational complexity. Dimensional reduction is a method to eliminate irrelevant features to improve the prediction accuracy. Typically, the eigenvalues or dimensional data variance from principal component analysis are used as criteria to select relevant features. This approach is simple but not efficient since it does not concern the degree of data overlap in each dimension in the feature space. A new method to select relevant features based on degree of dimensional data overlap with proper feature selection was introduced. Furthermore, our study concentrated on small sized data sets which usually occur in reality. The experimental results signified that this new approach can achieve substantially higher prediction accuracy when compared with other methods.

Extracting predictive SNPs in Crohn's disease using a vacillating genetic algorithm and a neural classifier in case-control association studies.

Crohn's disease is an inflammatory bowel disease. Because of strong heritability, it is possible to deploy the pattern of DNA variations, such as single nucleotide polymorphisms (SNPs), to accurately predict the state of this disease. However, there are many possible SNP subsets, which make finding a best set of SNPs to achieve the highest prediction accuracy impossible in one patient's lifetime. In this paper, a new technique is proposed that relies on chromosomes of various lengths with significant order feature selection, a new cross-over approach, and new mutation operations. Our method can find a chromosome of appropriate length with useful features. The Crohn's disease data that were gathered from case-control association studies were used to demonstrate the effectiveness of our proposed algorithm. In terms of the prediction accuracy, the proposed SNP prediction framework outperformed previously proposed techniques, including the optimum random forest (ORF), the univariate marginal distribution algorithm and support vector machine (USVM), the complimentary greedy search-based prediction algorithm (CGSP), the combinatorial search-based prediction algorithm (CSP), and discretized network flow (DNF). The performance of our framework, when tested against this real data set with a 5-fold cross-validation, was 90.4% accuracy with 87.5% sensitivity and 92.2% specificity.

Clinical prediction based on HPV DNA testing by hybrid capture 2 (HC2) in combination with liquid-based cytology (LBC).

Primary screening by HPV DNA testing is an effective method for reducing cervical cancer and has proven more sensitive than cytology. To advance this approach, many molecular methods have been developed. Hybrid capture 2 provides semi-quantitative results in ratios of relative light units and positive cutoff values (RLU/ PC). Twenty-five thousand and five patients were included in this study to analyze the correlation between the ratio of RLU/PC and stage of cervical dysplasia. The results show that the RLU/PC ratios ranged from 0-3500 while almost normal cases, ASC-US and ASC-H, had values below 200. Of those samples negative for cytology markers, 94.6% were normal and their RLU/PC ratios were less than 4. With an RLU/PC ratio greater than 4 and less than or equal to 300, the percentages in all age groups were normal 53.6%, LSIL 20.2%, ASC-US 17.2%, HSIL 6.13%, ASC-H 2.72%, and AGC 0.11%, respectively. In contrast, 64.0% of samples with a RLU/ PC ratio greater than 300 and less than or equal to 3500 were LSIL. These results should contribute to cost effective cervical cancer management strategies. Further studies of associations with particular HPV genotypes would be useful to predict the risk of progression to cancer.