A comparative evaluation of sequence classification programs

Background

Here we present two new computer tools, PREMIM and EMIM, for the estimation of parental and child genetic effects, based on genotype data from a variety of different child-parent configurations. PREMIM allows the extraction of child-parent genotype data from standard-format pedigree data files, while EMIM uses the extracted genotype data to perform subsequent statistical analysis. The use of genotype data from the parents as well as from the child in question allows the estimation of complex genetic effects such as maternal genotype effects, maternal-foetal interactions and parent-of-origin (imprinting) effects. These effects are estimated by EMIM, incorporating chosen assumptions such as Hardy-Weinberg equilibrium or exchangeability of parental matings as required.

Results

In application to simulated data, we show that the inference provided by EMIM is essentially equivalent to that provided by alternative (competing) software packages such as MENDEL and LEM. However, PREMIM and EMIM (used in combination) considerably outperform MENDEL and LEM in terms of speed and ease of execution.

Conclusions

Together, EMIM and PREMIM provide easy-to-use command-line tools for the analysis of pedigree data, giving unbiased estimates of parental and child genotype relative risks.     

Methods for prediction of peptide binding to MHC molecules: a comparative study

BACKGROUND: A variety of methods for prediction of peptide binding to major histocompatibility complex (MHC) have been proposed. These methods are based on binding motifs, binding matrices, hidden Markov models (HMM), or artificial neural networks (ANN). There has been little prior work on the comparative analysis of these methods. MATERIALS AND METHODS: We performed a comparison of the performance of six methods applied to the prediction of two human MHC class I molecules, including binding matrices and motifs, ANNs, and HMMs. RESULTS: The selection of the optimal prediction method depends on the amount of available data (the number of peptides of known binding affinity to the MHC molecule of interest), the biases in the data set and the intended purpose of the prediction (screening of a single protein versus mass screening). When little or no peptide data are available, binding motifs are the most useful alternative to random guessing or use of a complete overlapping set of peptides for selection of candidate binders. As the number of known peptide binders increases, binding matrices and HMM become more useful predictors. ANN and HMM are the predictive methods of choice for MHC alleles with more than 100 known binding peptides. CONCLUSION: The ability of bioinformatic methods to reliably predict MHC binding peptides, and thereby potential T-cell epitopes, has major implications for clinical immunology, particularly in the area of vaccine design.     

Feature ranking and rank aggregation for automatic sleep stage classification: a comparative study

Background

Nowadays, sleep quality is one of the most important measures of healthy life, especially considering the huge number of sleep-related disorders. Identifying sleep stages using polysomnographic (PSG) signals is the traditional way of assessing sleep quality. However, the manual process of sleep stage classification is time-consuming, subjective and costly. Therefore, in order to improve the accuracy and efficiency of the sleep stage classification, researchers have been trying to develop automatic classification algorithms. Automatic sleep stage classification mainly consists of three steps: pre-processing, feature extraction and classification. Since classification accuracy is deeply affected by the extracted features, a poor feature vector will adversely affect the classifier and eventually lead to low classification accuracy. Therefore, special attention should be given to the feature extraction and selection process.

Methods

In this paper the performance of seven feature selection methods, as well as two feature rank aggregation methods, were compared. Pz-Oz EEG, horizontal EOG and submental chin EMG recordings of 22 healthy males and females were used. A comprehensive feature set including 49 features was extracted from these recordings. The extracted features are among the most common and effective features used in sleep stage classification from temporal, spectral, entropy-based and nonlinear categories. The feature selection methods were evaluated and compared using three criteria: classification accuracy, stability, and similarity.

Results

Simulation results show that MRMR-MID achieves the highest classification performance while Fisher method provides the most stable ranking. In our simulations, the performance of the aggregation methods was in the average level, although they are known to generate more stable results and better accuracy.

Conclusions

The Borda and RRA rank aggregation methods could not outperform significantly the conventional feature ranking methods. Among conventional methods, some of them slightly performed better than others, although the choice of a suitable technique is dependent on the computational complexity and accuracy requirements of the user.

     

Peptidergic modulation of sleep: a comparative study of analogs of the peptide DSIP

Effects of intracerebroventricular administration of 3 DSIP analogues with higher stability against proteolysis, on subsequent sleep were studied in rabbits and rats previously implanted with electrodes and cannulas. Significant increase of total sleep time (mainly due to slow wave sleep) after administration of the peptides (D-Trp1) DSIP and (D-Tyr1) DSIP as compared with the control injections to the same animals was found during the first 3-5 hours in rabbits and 12 hours in rats. Shortened analogue (D-Trp1) DSIP1-6 had no effect on rabbit sleep and significantly reduced slow wave sleep in rats during the first 3 hours. It is concluded that some DSIP-like peptides which are more stable against aminopeptidases, can modulate rodent sleep.     

Pinacyanol as effective probe of fibrillar beta-amyloid peptide: comparative study with Congo Red

The binding of pinacyanol (PIN), a cationic cyanine dye, to beta-amyloid fibrils (Abeta), which are associated with Alzheimer disease, was quantified by absorption spectrophotometry to measure the concentration of PIN bound to Abeta as a function of the Abeta concentration or by means of the separation of free PIN from bound PIN by centrifugation and subsequent analysis of the supernatant by visible-absorption spectrophotometry. Both methods gave equivalent results. The stoichiometry of PIN binding to Abeta was 1, and the curve representing the concentration effect of Abeta on the concentration of a dye-Abeta complex showed a biphasic curve instead of the hyperbolic curve that is characteristic of weak ligand-macromolecule interactions [e.g., as shown by Congo Red (CR)]). This and the fact that a Scatchard plot could not be fitted to the experimental data suggested that PIN binds tightly to Abeta. A comparison to the interaction of CR with Abeta led us to conclude that PIN is more sensitive than CR.     

Pepsin as a catalyst of peptide synthesis. Enzyme co-precipitation with emerging peptide products.

Pepsin successfully catalyzed the synthesis of several peptide derivatives from N-protected di- or tripeptides and amino acid or peptide esters or p-nitroanilides in dimethylformamide-water solutions at pH 4.6. An optimal substrates:pepsin ratio depended on the structure of starting peptides, especially their fit to the substrate binding sites of the enzyme. For hexapeptide Z-Ala-Ala-Phe-Leu-Ala-Ala-OCH3 formation, an equilibrium yield was attained at 1:3.10(5) enzyme-substrates ratio that indicated high efficiency of pepsin in synthesis reactions. In the course of the equilibrium peptide synthesis, pepsin gradually disappeared from the liquid phase due to its entrapment within a gel, formed by the hexapeptide product, while retaining its activity. The inclusion into the precipitate was not specific for pepsin, so far as inert proteins, lysozyme, ribonuclease A and carbonic anhydrase, when added to the reaction mixture, became also co-precipitated with the hexapeptide formed. It appears that co-precipitation of pepsin, an important factor limiting the enzyme efficiency, might be operative as well for other proteinases used to catalyze peptide synthesis.     

Fault diagnosis of a benchmark fermentation process: a comparative study of feature extraction and classification techniques

This paper investigates fault diagnosis in batch processes and presents a comparative study of feature extraction and classification techniques applied to a specific biotechnological case study: the fermentation process model by Birol et al. (Comput Chem Eng 26:1553–1565, 2002), which is a benchmark for advanced batch processes monitoring, diagnosis and control. Fault diagnosis is achieved using four approaches on four different process scenarios based on the different levels of noise so as to evaluate their effects on the performance. Each approach combines a feature extraction method, either multi-way principal component analysis (MPCA) or multi-way independent component analysis (MICA), with a classification method, either artificial neural network (ANN) or support vector machines (SVM). The performance obtained by the different approaches is assessed and discussed for a set of simulated faults under different scenarios. One of the faults (a loss in mixing power) could not be detected due to the minimal effect of mixing on the simulated data. The remaining faults could be easily diagnosed and the subsequent discussion provides practical insight into the selection and use of the available techniques to specific applications. Irrespective of the classification algorithm, MPCA renders better results than MICA, hence the diagnosis performance proves to be more sensitive to the selection of the feature extraction technique.     

Racial classification in the evolutionary sciences: a comparative analysis

Human racial classification has long been a problem for the discipline of anthropology, but much of the criticism of the race concept has focused on its social and political connotations. The central argument of this paper is that race is not a specifically human problem, but one that exists in evolutionary thought in general. This paper looks at various disciplinary approaches to racial or subspecies classification, extending its focus beyond the anthropological race concept by providing a comparative analysis of the use of racial classification in evolutionary biology, genetics, and anthropology.     

Fast conventional Fmoc solid‐phase peptide synthesis: a comparative study of different activators

The ability to speed up conventional Fmoc solid‐phase peptide synthesis (SPPS) has many advantages including increased productivity. One way to speed up conventional Fmoc SPPS is the choice of activator. Recently, several new activators have been introduced into the market, and they were evaluated along with some older activators for their ability to synthesize a range of peptides with shorter and longer reaction times. It was found that HDMC, PyClock, COMU, HCTU, and HATU worked well at shorter reaction times (2 × 1 min), but PyOxim and TFFH only worked well at longer reaction times. The performance of PyBOP at shorter reaction times was poor only for more difficult sequences. These results are important for selecting an appropriate activator for fast SPPS applications. Copyright © 2011 European Peptide Society and John Wiley & Sons, Ltd.     

Enzyme histochemical characteristics of human and kitten odontoclasts and kitten osteoclasts: a comparative study using whole cells

Synopsis Methods for the histochemical demonstration of enzymes in whole cell preparations of odontoclasts and osteoclasts are described. Enzyme histochemical characteristics of human and kitten odontoclasts from resorbing primary teeth and of osteoclasts from kitten femur metaphyses were determined and compared. The enzyme profiles, times for the appearance of detectable reaction product, intensity of the reactions and localization of the reaction products were similar in all three types of giant cell. These findings suggest that odontoclasts have enzyme properties and metabolic functions similar to those of osteoclasts. Species differences appear to be minor, although the NADP-dependent enzymes are less active in human than in kitten odontoclasts. Both odontoclasts and osteoclasts are rich in enzymes concerned with energy production and possess considerable activity of enzymes usually associated with catabolic functions. Metabolic pathways are well developed in respect of the utilization of succinic, malic, glutamic, lactic and isocitric acids, -hydroxybutyric acid and glucose-6-phosphate, and they also possess phosphatases, non-specific esterases and leucine naphthylamidase. The distribution of enzyme reaction products for the individual enzymes demonstrated is consistent with the presence in these cells of large numbers of mitochondria and lysosome-like organelles. Considerable phosphatase activity is demonstrable in both odontoclasts and osteoclasts at both neutral and acid pH.     

Enzyme classification by ligand binding

The problem of assigning a biochemical function to newly discovered proteins has been traditionally approached by expert enzymological analysis, sequence analysis, and structural modeling. In recent years, the appearance of databases containing protein-ligand interaction data for large numbers of protein classes and chemical compounds have provided new ways of investigating proteins for which the biochemical function is not completely understood. In this work, we introduce a method that utilizes ligand-binding data for functional classification of enzymes. The method makes use of the existing Enzyme Commission (EC) classification scheme and the data on interactions of small molecules with enzymes from the BRENDA database. A set of ligands that binds to an enzyme with unknown biochemical function serves as a query to search a protein-ligand interaction database for enzyme classes that are known to interact with a similar set of ligands. These classes provide hypotheses of the query enzyme's function and complement other computational annotations that take advantage of sequence and structural information. Similarity between sets of ligands is computed using point set similarity measures based upon similarity between individual compounds. We present the statistics of classification of the enzymes in the database by a cross-validation procedure and illustrate the application of the method on several examples.     

Rapid screening: a comparative study

Although rapid screening of negative and inadequate cervical smears is a quality assurance requirement for all UK laboratories, there has been little attempt to standardize the method and laboratories make use of a number of different techniques and times. The aim of this study was to assess the sensitivity of these various techniques by measuring their ability to pick out known false-negative smears. Completed questionnaires from 123 laboratories across England revealed that 52% of laboratories use a "step" technique, 19% use "turret", 15% use random paths and 34% attempt to rescreen the whole slide quickly. Twenty-two percent of laboratories use a mixture of techniques. Timings are also variable, with the majority of laboratories allowing screeners to review slides at a pace decided by themselves but usually between 1 and 2 min. The study involved 120 participants who performed a total of 24 000 rapid screens. The results showed that, of the 90 abnormal slides used in the study, 62 cases (69%) were identified as abnormal or needing review by more than 50% of participants. Overall rapid screening picked out 58% of high-grade squamous abnormalities, 59% of low-grade abnormalities and 72% of glandular lesions. Step screening performed best, followed by whole slide/random and then turret. One minute was the optimum time and there was a significant fall in performance once individuals attempted to rescreen large numbers (>50). The most significant finding was the marked variation in the performance of individuals using the same slide sets.     

Sample size and statistical power considerations in high-dimensionality data settings: a comparative study of classification algorithms

Background  

Data generated using 'omics' technologies are characterized by high dimensionality, where the number of features measured per subject vastly exceeds the number of subjects in the study. In this paper, we consider issues relevant in the design of biomedical studies in which the goal is the discovery of a subset of features and an associated algorithm that can predict a binary outcome, such as disease status. We compare the performance of four commonly used classifiers (K-Nearest Neighbors, Prediction Analysis for Microarrays, Random Forests and Support Vector Machines) in high-dimensionality data settings. We evaluate the effects of varying levels of signal-to-noise ratio in the dataset, imbalance in class distribution and choice of metric for quantifying performance of the classifier. To guide study design, we present a summary of the key characteristics of 'omics' data profiled in several human or animal model experiments utilizing high-content mass spectrometry and multiplexed immunoassay based techniques.     

Enzymatic peptide synthesis in organic media: a comparative study of water-miscible and water-immiscible solvent systems.

Peptide synthesis was carried out in a variety of organic solvents with low contents of water. The enzyme was deposited on the support material, celite, from an aqueous buffer solution. After evaporation of the water the biocatalyst was suspended in the reaction mixtures. The chymotrypsin-catalyzed reaction between Z-Phe-OMe and Leu-NH2 was used as a model reaction. Under the conditions used ([Z-Phe-OMe]0 less than or equal to 40 mM, [Leu-NH2]0/([Z-Phe-OMe]0 = 1.5) the reaction was first order with respect to Z-Phe-OMe. Tris buffer, pH 7.8, was the best buffer to use in the preparation of the biocatalyst. In water-miscible solvents the reaction rate increased with increasing water content, but the final yield of peptide decreased due to the competing hydrolysis of Z-Phe-OMe. Among the water-miscible solvents, acetonitrile was the most suitable, giving 91% yield with 4% (by vol.) water. In water-immiscible solvents the reaction rate and the product distribution were little affected by water additions in the range between 0% and 2% (vol. %) in excess of water saturation. The reaction rates correlated well with the log P values of the solvent. The highest yield (93%) was obtained in ethyl acetate; in this solvent the reaction was also fast. Under most reaction conditions used the reaction product was stable; secondary hydrolysis of the peptide formed was normally negligible. The method presented is a combination of kinetically controlled peptide synthesis (giving high reaction rates) and thermodynamically controlled peptide synthesis (giving stable reaction products).     

Computational comparative study of tuberculosis proteomes using a model learned from signal peptide structures

Secretome analysis is important in pathogen studies. A fundamental and convenient way to identify secreted proteins is to first predict signal peptides, which are essential for protein secretion. However, signal peptides are highly complex functional sequences that are easily confused with transmembrane domains. Such confusion would obviously affect the discovery of secreted proteins. Transmembrane proteins are important drug targets, but very few transmembrane protein structures have been determined experimentally; hence, prediction of the structures is essential. In the field of structure prediction, researchers do not make assumptions about organisms, so there is a need for a general signal peptide predictor.To improve signal peptide prediction without prior knowledge of the associated organisms, we present a machine-learning method, called SVMSignal, which uses biochemical properties as features, as well as features acquired from a novel encoding, to capture biochemical profile patterns for learning the structures of signal peptides directly.We tested SVMSignal and five popular methods on two benchmark datasets from the SPdb and UniProt/Swiss-Prot databases, respectively. Although SVMSignal was trained on an old dataset, it performed well, and the results demonstrate that learning the structures of signal peptides directly is a promising approach. We also utilized SVMSignal to analyze proteomes in the entire HAMAP microbial database. Finally, we conducted a comparative study of secretome analysis on seven tuberculosis-related strains selected from the HAMAP database. We identified ten potential secreted proteins, two of which are drug resistant and four are potential transmembrane proteins.SVMSignal is publicly available at http://bio-cluster.iis.sinica.edu.tw/SVMSignal. It provides user-friendly interfaces and visualizations, and the prediction results are available for download.     

Bayesian classification and non-Bayesian label estimation via EM algorithm to identify differentially expressed genes: a comparative study

Gene classification problem is studied considering the ratio of gene expression levels, X, in two-channel microarrays and a non-observed categorical variable indicating how differentially expressed the gene is: non differentially expressed, down-regulated or up-regulated. Supposing X from a mixture of Gamma distributions, two methods are proposed and results are compared. The first method is based on an hierarchical Bayesian model. The conditional predictive probability of a gene to belong to each group is calculated and the gene is assigned to the group for which this conditional probability is higher. The second method uses EM algorithm to estimate the most likely group label for each gene, that is, to assign the gene to the group which contains it with the higher estimated probability.     

The sarcoplasmic reticulum: a comparative study

The diversity of cellular membrane structures associated with regulation of intracellular calcium level is described in several different groups of organisms and cells. All the instances reported refer to cellular processes related to movement, in which calcium ion acts as trigger and/or modulator. In addition, a simplified five-stage picture of the underlying view of evolution of these structures is presented. In short: the choice made by nature in using calcium as intracellular messenger was very early in the history of life; all cellular structures devoted to intracellular calcium regulation, from the simplest form of amoeba to the highly sophisticated apparatus of mammalian skeletal muscle, can be linked together in the chain of evolution. Because the evidence is still sparse, any conclusion more positive would be speculative and of little value. Hopefully, in the coming years, with a better understanding of membrane architecture as a whole and its protein components (i.e. calcium channels, calcium-binding proteins), we will be able to test the first segments of this evolutionary hypothesis.