HMMGEP: clustering gene expression data using hidden Markov models

SUMMARY: The package HMMGEP performs cluster analysis on gene expression data using hidden Markov models. AVAILABILITY: HMMGEP, including the source code, documentation and sample data files, is available at http://www.bioinfo.tsinghua.edu.cn:8080/~rich/hmmgep_download/index.html.     

Using hidden Markov models to analyze gene expression time course data

MOTIVATION: Cellular processes cause changes over time. Observing and measuring those changes over time allows insights into the how and why of regulation. The experimental platform for doing the appropriate large-scale experiments to obtain time-courses of expression levels is provided by microarray technology. However, the proper way of analyzing the resulting time course data is still very much an issue under investigation. The inherent time dependencies in the data suggest that clustering techniques which reflect those dependencies yield improved performance. RESULTS: We propose to use Hidden Markov Models (HMMs) to account for the horizontal dependencies along the time axis in time course data and to cope with the prevalent errors and missing values. The HMMs are used within a model-based clustering framework. We are given a number of clusters, each represented by one Hidden Markov Model from a finite collection encompassing typical qualitative behavior. Then, our method finds in an iterative procedure cluster models and an assignment of data points to these models that maximizes the joint likelihood of clustering and models. Partially supervised learning--adding groups of labeled data to the initial collection of clusters--is supported. A graphical user interface allows querying an expression profile dataset for time course similar to a prototype graphically defined as a sequence of levels and durations. We also propose a heuristic approach to automate determination of the number of clusters. We evaluate the method on published yeast cell cycle and fibroblasts serum response datasets, and compare them, with favorable results, to the autoregressive curves method.     

Mining Bacillus subtilis chromosome heterogeneities using hidden Markov models

We present here the use of a new statistical segmentation method on the Bacillus subtilis chromosome sequence. Maximum likelihood parameter estimation of a hidden Markov model, based on the expectation-maximization algorithm, enables one to segment the DNA sequence according to its local composition. This approach is not based on sliding windows; it enables different compositional classes to be separated without prior knowledge of their content, size and localization. We compared these compositional classes, obtained from the sequence, with the annotated DNA physical map, sequence homologies and repeat regions. The first heterogeneity revealed discriminates between the two coding strands and the non-coding regions. Other main heterogeneities arise; some are related to horizontal gene transfer, some to t-enriched composition of hydrophobic protein coding strands, and others to the codon usage fitness of highly expressed genes. Concerning potential and established gene transfers, we found 9 of the 10 known prophages, plus 14 new regions of atypical composition. Some of them are surrounded by repeats, most of their genes have unknown function or possess homology to genes involved in secondary catabolism, metal and antibiotic resistance. Surprisingly, we notice that all of these detected regions are a + t-richer than the host genome, raising the question of their remote sources.     

The analysis of hospital infection data using hidden Markov models

Surveillance data for communicable nosocomial pathogens usually consist of short time series of low-numbered counts of infected patients. These often show overdispersion and autocorrelation. To date, almost all analyses of such data have ignored the communicable nature of the organisms and have used methods appropriate only for independent outcomes. Inferences that depend on such analyses cannot be considered reliable when patient-to-patient transmission is important. We propose a new method for analysing these data based on a mechanistic model of the epidemic process. Since important nosocomial pathogens are often carried asymptomatically with overt infection developing in only a proportion of patients, the epidemic process is usually only partially observed by routine surveillance data. We therefore develop a 'structured' hidden Markov model where the underlying Markov chain is generated by a simple transmission model. We apply both structured and standard (unstructured) hidden Markov models to time series for three important pathogens. We find that both methods can offer marked improvements over currently used approaches when nosocomial spread is important. Compared to the standard hidden Markov model, the new approach is more parsimonious, is more biologically plausible, and allows key epidemiological parameters to be estimated.     

Mining DNA microarray data using a novel approach based on graph theory.

The recent demonstration that biochemical pathways from diverse organisms are arranged in scale-free, rather than random, systems [Jeong et al., Nature 407 (2000) 651-654], emphasizes the importance of developing methods for the identification of biochemical nexuses--the nodes within biochemical pathways that serve as the major input/output hubs, and therefore represent potentially important targets for modulation. Here we describe a bioinformatics approach that identifies candidate nexuses for biochemical pathways without requiring functional gene annotation; we also provide proof-of-principle experiments to support this technique. This approach, called Nexxus, may lead to the identification of new signal transduction pathways and targets for drug design.     

Mining gene expression data based on template theory

MOTIVATION: It is understood that clustering genes are useful for exploring scientific knowledge from DNA microarray gene expression data. The explored knowledge can be finally used for annotating biological function for novel genes. Representing the explored knowledge in an efficient manner is then closely related to the classification accuracy. However, this issue has not yet been paid the attention it deserves. RESULT: A novel method based on template theory in cognitive psychology and pattern recognition is developed in this study for representing knowledge extracted from cluster analysis effectively. The basic principle is to represent knowledge according to the relationship between genes and a found cluster structure. Based on this novel knowledge representation method, a pattern recognition algorithm (the decision tree algorithm C4.5) is then used to construct a classifier for annotating biological functions of novel genes. The experiments on five published datasets show that this method has improved the classification performance compared with the conventional method. The statistical tests indicate that this improvement is significant. AVAILABILITY: The software package can be obtained upon request from the author.     

Calibrating E-values for hidden Markov models using reverse-sequence null models

MOTIVATION: Hidden Markov models (HMMs) calculate the probability that a sequence was generated by a given model. Log-odds scoring provides a context for evaluating this probability, by considering it in relation to a null hypothesis. We have found that using a reverse-sequence null model effectively removes biases owing to sequence length and composition and reduces the number of false positives in a database search. Any scoring system is an arbitrary measure of the quality of database matches. Significance estimates of scores are essential, because they eliminate model- and method-dependent scaling factors, and because they quantify the importance of each match. Accurate computation of the significance of reverse-sequence null model scores presents a problem, because the scores do not fit the extreme-value (Gumbel) distribution commonly used to estimate HMM scores' significance. RESULTS: To get a better estimate of the significance of reverse-sequence null model scores, we derive a theoretical distribution based on the assumption of a Gumbel distribution for raw HMM scores and compare estimates based on this and other distribution families. We derive estimation methods for the parameters of the distributions based on maximum likelihood and on moment matching (least-squares fit for Student's t-distribution). We evaluate the modeled distributions of scores, based on how well they fit the tail of the observed distribution for data not used in the fitting and on the effects of the improved E-values on our HMM-based fold-recognition methods. The theoretical distribution provides some improvement in fitting the tail and in providing fewer false positives in the fold-recognition test. An ad hoc distribution based on assuming a stretched exponential tail does an even better job. The use of Student's t to model the distribution fits well in the middle of the distribution, but provides too heavy a tail. The moment-matching methods fit the tails better than maximum-likelihood methods. AVAILABILITY: Information on obtaining the SAM program suite (free for academic use), as well as a server interface, is available at http://www.soe.ucsc.edu/research/compbio/sam.html and the open-source random sequence generator with varying compositional biases is available at http://www.soe.ucsc.edu/research/compbio/gen_sequence     

An approach to determine myocardial ischemia by hidden Markov models

A hidden Markov model (HMM) of electrocardiogram (ECG) signal is presented for detection of myocardial ischemia. The time domain signals that are recorded by the ECG before and during the episode of local ischemia were pre-processed to produce input sequences, which is needed for the model training. The model is also verified by test data, and the results show that the models have certain function for the detection of myocardial ischemia. The algorithm based on HMM provides a possible approach for the timely, rapid and automatic diagnosis of myocardial ischemia, and also can be used in portable medical diagnostic equipment in the future.     

An approach to determine myocardial ischemia by hidden Markov models

A hidden Markov model (HMM) of electrocardiogram (ECG) signal is presented for detection of myocardial ischemia. The time domain signals that are recorded by the ECG before and during the episode of local ischemia were pre-processed to produce input sequences, which is needed for the model training. The model is also verified by test data, and the results show that the models have certain function for the detection of myocardial ischemia. The algorithm based on HMM provides a possible approach for the timely, rapid and automatic diagnosis of myocardial ischemia, and also can be used in portable medical diagnostic equipment in the future.     

A quantitative approach for polymerase chain reactions based on a hidden Markov model

Polymerase chain reaction (PCR) is a major DNA amplification technology from molecular biology. The quantitative analysis of PCR aims at determining the initial amount of the DNA molecules from the observation of typically several PCR amplifications curves. The mainstream observation scheme of the DNA amplification during PCR involves fluorescence intensity measurements. Under the classical assumption that the measured fluorescence intensity is proportional to the amount of present DNA molecules, and under the assumption that these measurements are corrupted by an additive Gaussian noise, we analyze a single amplification curve using a hidden Markov model(HMM). The unknown parameters of the HMM may be separated into two parts. On the one hand, the parameters from the amplification process are the initial number of the DNA molecules and the replication efficiency, which is the probability of one molecule to be duplicated. On the other hand, the parameters from the observational scheme are the scale parameter allowing to convert the fluorescence intensity into the number of DNA molecules and the mean and variance characterizing the Gaussian noise. We use the maximum likelihood estimation procedure to infer the unknown parameters of the model from the exponential phase of a single amplification curve, the main parameter of interest for quantitative PCR being the initial amount of the DNA molecules. An illustrative example is provided. This research was financed by the Swedish foundation for Strategic Research through the Gothenburg Mathematical Modelling Centre.     

Pair hidden Markov models on tree structures

MOTIVATION: Computationally identifying non-coding RNA regions on the genome has much scope for investigation and is essentially harder than gene-finding problems for protein-coding regions. Since comparative sequence analysis is effective for non-coding RNA detection, efficient computational methods are expected for structural alignments of RNA sequences. On the other hand, Hidden Markov Models (HMMs) have played important roles for modeling and analysing biological sequences. Especially, the concept of Pair HMMs (PHMMs) have been examined extensively as mathematical models for alignments and gene finding. RESULTS: We propose the pair HMMs on tree structures (PHMMTSs), which is an extension of PHMMs defined on alignments of trees and provides a unifying framework and an automata-theoretic model for alignments of trees, structural alignments and pair stochastic context-free grammars. By structural alignment, we mean a pairwise alignment to align an unfolded RNA sequence into an RNA sequence of known secondary structure. First, we extend the notion of PHMMs defined on alignments of 'linear' sequences to pair stochastic tree automata, called PHMMTSs, defined on alignments of 'trees'. The PHMMTSs provide various types of alignments of trees such as affine-gap alignments of trees and an automata-theoretic model for alignment of trees. Second, based on the observation that a secondary structure of RNA can be represented by a tree, we apply PHMMTSs to the problem of structural alignments of RNAs. We modify PHMMTSs so that it takes as input a pair of a 'linear' sequence and a 'tree' representing a secondary structure of RNA to produce a structural alignment. Further, the PHMMTSs with input of a pair of two linear sequences is mathematically equal to the pair stochastic context-free grammars. We demonstrate some computational experiments to show the effectiveness of our method for structural alignments, and discuss a complexity issue of PHMMTSs.     

Bayesian basecalling for DNA sequence analysis using hidden Markov models

It has been shown that electropherograms of DNA sequences can be modeled with hidden Markov models. Basecalling, the procedure that determines the sequence of bases from the given eletropherogram, can then be performed using the Viterbi algorithm. A training step is required prior to basecalling in order to estimate the HMM parameters. In this paper, we propose a Bayesian approach which employs the Markov chain Monte Carlo (MCMC) method to perform basecalling. Such an approach not only allows one to naturally encode the prior biological knowledge into the basecalling algorithm, it also exploits both the training data and the basecalling data in estimating the HMM parameters, leading to more accurate estimates. Using the recently sequenced genome of the organism Legionella pneumophila we show that the MCMC basecaller outperforms the state-of-the-art basecalling algorithm in terms of total errors while requiring much less training than other proposed statistical basecallers.     

Sequence alignments and pair hidden Markov models using evolutionary history

This work presents a novel pairwise statistical alignment method based on an explicit evolutionary model of insertions and deletions (indels). Indel events of any length are possible according to a geometric distribution. The geometric distribution parameter, the indel rate, and the evolutionary time are all maximum likelihood estimated from the sequences being aligned. Probability calculations are done using a pair hidden Markov model (HMM) with transition probabilities calculated from the indel parameters. Equations for the transition probabilities make the pair HMM closely approximate the specified indel model. The method provides an optimal alignment, its likelihood, the likelihood of all possible alignments, and the reliability of individual alignment regions. Human alpha and beta-hemoglobin sequences are aligned, as an illustration of the potential utility of this pair HMM approach.     

Bayesian restoration of ion channel records using hidden Markov models

Hidden Markov models have been used to restore recorded signals of single ion channels buried in background noise. Parameter estimation and signal restoration are usually carried out through likelihood maximization by using variants of the Baum-Welch forward-backward procedures. This paper presents an alternative approach for dealing with this inferential task. The inferences are made by using a combination of the framework provided by Bayesian statistics and numerical methods based on Markov chain Monte Carlo stochastic simulation. The reliability of this approach is tested by using synthetic signals of known characteristics. The expectations of the model parameters estimated here are close to those calculated using the Baum-Welch algorithm, but the present methods also yield estimates of their errors. Comparisons of the results of the Bayesian Markov Chain Monte Carlo approach with those obtained by filtering and thresholding demonstrate clearly the superiority of the new methods.