McTwo: a two-step feature selection algorithm based on maximal information coefficient.

BACKGROUND: High-throughput bio-OMIC technologies are producing high-dimension data from bio-samples at an ever increasing rate, whereas the training sample number in a traditional experiment remains small due to various difficulties. This "large p, small n" paradigm in the area of biomedical "big data" may be at least partly solved by feature selection algorithms, which select only features significantly associated with phenotypes. Feature selection is an NP-hard problem. Due to the exponentially increased time requirement for finding the globally optimal solution, all the existing feature selection algorithms employ heuristic rules to find locally optimal solutions, and their solutions achieve different performances on different datasets. RESULTS: This work describes a feature selection algorithm based on a recently published correlation measurement, Maximal Information Coefficient (MIC). The proposed algorithm, McTwo, aims to select features associated with phenotypes, independently of each other, and achieving high classification performance of the nearest neighbor algorithm. Based on the comparative study of 17 datasets, McTwo performs about as well as or better than existing algorithms, with significantly reduced numbers of selected features. The features selected by McTwo also appear to have particular biomedical relevance to the phenotypes from the literature. CONCLUSION: McTwo selects a feature subset with very good classification performance, as well as a small feature number. So McTwo may represent a complementary feature selection algorithm for the high-dimensional biomedical datasets.

Evolutionary optimization of transcription factor binding motif detection.

All the cell types are under strict control of how their genes are transcribed into expressed transcripts by the temporally dynamic orchestration of the transcription factor binding activities. Given a set of known binding sites (BSs) of a given transcription factor (TF), computational TFBS screening technique represents a cost efficient and large scale strategy to complement the experimental ones. There are two major classes of computational TFBS prediction algorithms based on the tertiary and primary structures, respectively. A tertiary structure based algorithm tries to calculate the binding affinity between a query DNA fragment and the tertiary structure of the given TF. Due to the limited number of available TF tertiary structures, primary structure based TFBS prediction algorithm is a necessary complementary technique for large scale TFBS screening. This study proposes a novel evolutionary algorithm to randomly mutate the weights of different positions in the binding motif of a TF, so that the overall TFBS prediction accuracy is optimized. The comparison with the most widely used algorithm, Position Weight Matrix (PWM), suggests that our algorithm performs better or the same level in all the performance measurements, including sensitivity, specificity, accuracy and Matthews correlation coefficient. Our data also suggests that it is necessary to remove the widely used assumption of independence between motif positions. The supplementary material may be found at: http://www.healthinformaticslab.org/supp/ .

Gene expression profile based classification models of psoriasis.

Psoriasis is an autoimmune disease, which symptoms can significantly impair the patient's life quality. It is mainly diagnosed through the visual inspection of the lesion skin by experienced dermatologists. Currently no cure for psoriasis is available due to limited knowledge about its pathogenesis and development mechanisms. Previous studies have profiled hundreds of differentially expressed genes related to psoriasis, however with no robust psoriasis prediction model available. This study integrated the knowledge of three feature selection algorithms that revealed 21 features belonging to 18 genes as candidate markers. The final psoriasis classification model was established using the novel Incremental Feature Selection algorithm that utilizes only 3 features from 2 unique genes, IGFL1 and C10orf99. This model has demonstrated highly stable prediction accuracy (averaged at 99.81%) over three independent validation strategies. The two marker genes, IGFL1 and C10orf99, were revealed as the upstream components of growth signal transduction pathway of psoriatic pathogenesis.

Characterization of bacterial diversity and screening of cellulose-degrading bacteria in the gut system of Glenea cantor (Fabricius) larvae.

The intestinal bacteria of longhorn beetles would be ideal targets for pest control and lignocellulosic resources by destroying or exploiting their cellulose-degrading function. This article aims to investigate the diversity and community structure of intestinal bacteria the oligophagous longhorn beetle Glenea cantor. Additionally, it seeks to identify the presence of lignocellulose-degrading bacteria in the gut, and explore their role in consuming host kapok trees Bombax malabaricum. In this study, the bacterial community from G. cantor was examined by Illumina sequencing of 16S ribosomal RNA (rRNA) targeting the V3 and V4 regions. A total of 563,201 valid sequences and 814 OTUs were obtained. The dominant phyla were Proteobacteria, and the dominant genera were Acinetobacter and Lactococcus. The analysis of microbial diversity revealed a high bacterial diversity in the samples, with the gut bacteria playing a crucial role in the physiological activities of the host, particularly, 9 genera of intestinal bacteria with cellulose degradation function were found, highlighting their vital role in cellulose degradation. Five strains of cellulose-degrading bacteria, belonging to the genus Pseudomonas, were obtained from the intestinal tract of G. cantor larvae using traditional isolation and culture techniques as well as 16S rDNA sequencing. Among these strains, A4 exhibited a cellulase activity of 94.42 ± 0.42 U/mL, while A5 displayed the highest filter paper enzyme activity of 127.46 ± 3.54 U/mL. These results offered valuable insights into potential targets for pest control through internal attack digestion and cellulose-degrading bacteria in longhorn beetles.

Clinical effects of Xinmailong therapy in patients with chronic heart failure.

In the last 100 years, intensive studies have been done on the identification of the systematic approaches to find the cure for the chronic heart failure, however the mystery remains unresolved due to its complicated pathogenesis and ineffective early diagnosis. The present investigation was aimed to evaluate the potential effects of the traditional chinese medicine, Xinmailong, on the chronic heart failure (CHF) patients as compared to the standard western medical treatment available so far. In our study, we selected two groups of voluntary CHF patients at the Xiangya Hospital, which were allowed to administrate Xinmailong or standard treatments, respectively. Another group of voluntary healthy individuals were recruited as the control group. The treatment effectiveness was measured by five symptomatic factors, i.e. angiotensin II (Ang_II), high sensitivity C-reactive protein (hs_CRP), Left Ventricular End Systolic Volume Index (LVESVI), left ventricular ejection fraction (LVEF) and pro-B-type natriuretic peptide (NT_proBNP), between the control group and the CHF patients at different stages of drug administration and in different treatment groups. The timeline for the full dose administration was set to 15 days and five measurements as indicated above were taken on every 0, 7th and 15th day of the drug administration respectively. In the conducted study, similar symptomatic measurements were observed on day 0 in both treatment groups, and slight improvements were observed on 7th day. It was observed that after a full course of drug administration for 15 days, both of the treatment groups achieved statistically significant improvements in all the five measures, but Xinmailong was found to be more (almost double) statistically significant as compared with the available drug treatments for chronic heart failure.

Does self-medication reduce medical expenditure among the middle-aged and elderly population? A four-wave longitudinal study in China.

Introduction: Self-medication has a high prevalence in the middle-aged and elderly population in China. Despite the published evidence demonstrating the economic benefits of self-medication, limited research has addressed the relationship between self-medication and individual medical expenditures, especially within the Chinese population. This study examined the effect of self-medication on individual medical expenditures in China and analyzed the heterogeneity between outpatient and inpatient cases. Methods: We conducted a panel data analysis using data from four waves of the China Health and Retirement Longitudinal Study (CHARLS). Two-part mixed-effect models were implemented to estimate the effect of self-medication on total outpatient and inpatient expenses and out-of-pocket (OOP) costs, where mixed-effects logit regression was used as the first part, and generalized linear mixed models with log link and gamma distribution was used as the second part. Results: We identified 72,041 responses representing 24,641 individuals, of which 13,185 responses incurred outpatient expenses and 9,003 responses incurred inpatient costs. Controlling for all covariates, we found that self-medication behaviors were significantly associated with a higher probability of outpatient service utilization (OR = 1.250, 95% CI = 0.179 to 0.269; P < 0.001), but displayed no significant association with outpatient expenses. Respondents who had taken self-medication were less likely to use inpatient services (OR = 0.865, 95% CI = -0.201 to -0.089; P < 0.001), and their inpatient expenses were significantly reduced by 9.4% (P < 0.001). Inpatient OOP costs were significantly reduced by 10.7% (P < 0.001), and outpatient OOP costs were significantly increased by 11.3% (P < 0.001) among respondents who had self-medicated. Conclusions: This study allowed us to identify the economic value of self-medication among the middle-aged and elderly population in China. Future work should guide the middle-aged and elderly to take responsible self-medication to reduce their economic burden.

CRISPRdigger: detecting CRISPRs with better direct repeat annotations.

Clustered regularly interspaced short palindromic repeats (CRISPRs) are important genetic elements in many bacterial and archaeal genomes, and play a key role in prokaryote immune systems' fight against invasive foreign elements. The CRISPR system has also been engineered to facilitate target gene editing in eukaryotic genomes. Using the common features of mis-annotated CRISPRs in prokaryotic genomes, this study proposed an accurate de novo CRISPR annotation program CRISPRdigger, which can take a partially assembled genome as its input. A comprehensive comparison with the three existing programs demonstrated that CRISPRdigger can recover more Direct Repeats (DRs) for CRISPRs and achieve a higher accuracy for a query genome. The program was implemented by Perl and all the parameters had default values, so that a user could annotate CRISPRs in a query genome by supplying only a genome sequence in the FASTA format. All the supplementary data are available at http://www.healthinformaticslab.org/supp/.

jEcho: an Evolved weight vector to CHaracterize the protein's posttranslational modification mOtifs.

Protein's posttranslational modification (PTM) represents a major dynamic regulation of protein functions after the translation of polypeptide chains from mRNA molecule. Compared with the costly and labor-intensive wet laboratory characterization of PTMs, the computer-based detection of PTM residues has been a major complementary technique in recent years. Previous studies demonstrated that the PTM-flanking positions convey different contributions to the computational detection of PTM residue, but did not directly translate this observation into the in silico PTM prediction. We propose a weight vector to represent the variant contributions of the PTM-flanking positions and use an evolutionary algorithm to optimize the vector. Even a simple nearest neighbor algorithm with the incorporated optimal weight vector outperforms the currently available algorithms. The algorithm is implemented as an easy-to-use computer program, jEcho version 1.0. The implementation language, Java, makes jEcho platform-independent and visually interactive. The predicted results may be directly exported as publication-quality images or text files. jEcho may be downloaded from http://www.healthinformaticslab.org/supp/ .

jEcho: an evolved weight vector to characterize the protein's post-translational modification motifs.

Protein's post-translational modification (PTM) represents a major dynamic regulation of protein functions after the translation of polypeptide chains from mRNA molecule. Compared with the costly and labor intensive wet lab characterization of PTMs, the computer-based detection of PTM residues has been a major complementary technique in recent years. Previous studies demonstrated that the PTM-flanking positions convey different contributions to the computational detection of PTM residue, but did not directly translate this observation into the in silico PTM prediction. We propose a weight vector to represent the variant contributions of the PTM flanking positions, and use an evolutionary algorithm to optimize the vector. Even a simple nearest neighbor algorithm with the incorporated optimal weight vector outperforms the currently available algorithms. The algorithm is implemented as an easy-to-use computer program, jEcho version 1.0. The implementation language, Java, makes jEcho platform-independent and visually interactive. The predicted results may be directly exported as publication-quality images or text files. jEcho may be downloaded from http://www.healthinformaticslab.org/supp/ .

Constraint programming based biomarker optimization.

Efficient and intuitive characterization of biological big data is becoming a major challenge for modern bio-OMIC based scientists. Interactive visualization and exploration of big data is proven to be one of the successful solutions. Most of the existing feature selection algorithms do not allow the interactive inputs from users in the optimizing process of feature selection. This study investigates this question as fixing a few user-input features in the finally selected feature subset and formulates these user-input features as constraints for a programming model. The proposed algorithm, fsCoP (feature selection based on constrained programming), performs well similar to or much better than the existing feature selection algorithms, even with the constraints from both literature and the existing algorithms. An fsCoP biomarker may be intriguing for further wet lab validation, since it satisfies both the classification optimization function and the biomedical knowledge. fsCoP may also be used for the interactive exploration of bio-OMIC big data by interactively adding user-defined constraints for modeling.

A novel electrocardiogram parameterization algorithm and its application in myocardial infarction detection.

The electrocardiogram (ECG) is a biophysical electric signal generated by the heart muscle, and is one of the major measurements of how well a heart functions. Automatic ECG analysis algorithms usually extract the geometric or frequency-domain features of the ECG signals and have already significantly facilitated automatic ECG-based cardiac disease diagnosis. We propose a novel ECG feature by fitting a given ECG signal with a 20th order polynomial function, defined as PolyECG-S. The PolyECG-S feature is almost identical to the fitted ECG curve, measured by the Akaike information criterion (AIC), and achieved a 94.4% accuracy in detecting the Myocardial Infarction (MI) on the test dataset. Currently ST segment elongation is one of the major ways to detect MI (ST-elevation myocardial infarction, STEMI). However, many ECG signals have weak or even undetectable ST segments. Since PolyECG-S does not rely on the information of ST waves, it can be used as a complementary MI detection algorithm with the STEMI strategy. Overall, our results suggest that the PolyECG-S feature may satisfactorily reconstruct the fitted ECG curve, and is complementary to the existing ECG features for automatic cardiac function analysis.