Uday Kamath

An Evolutionary Algorithm Approach for Feature Generation from Sequence Data and Its Application to DNA Splice Site Prediction

Associating functional information with biological sequences remains a challenge for machine learning methods. The performance of these methods often depends on deriving predictive features from the sequences sought to be classified. Feature generation is a difficult problem, as the connection between the sequence features and the sought property is not known a priori. It is often the task of domain experts or exhaustive feature enumeration techniques to generate a few features whose predictive power is then tested in the context of classification. This paper proposes an evolutionary algorithm to effectively explore a large feature space and generate predictive features from sequence data. The effectiveness of the algorithm is demonstrated on an important component of the gene-finding problem, DNA splice site prediction. This application is chosen due to the complexity of the features needed to obtain high classification accuracy and precision. Our results test the effectiveness of the obtained features in the context of classification by Support Vector Machines and show significant improvement in accuracy and precision over state-of-the-art approaches.

Effective Automated Feature Construction and Selection for Classification of Biological Sequences

Background Many open problems in bioinformatics involve elucidating underlying functional signals in biological sequences. DNA sequences, in particular, are characterized by rich architectures in which functional signals are increasingly found to combine local and distal interactions at the nucleotide level. Problems of interest include detection of regulatory regions, splice sites, exons, hypersensitive sites, and more. These problems naturally lend themselves to formulation as classification problems in machine learning. When classification is based on features extracted from the sequences under investigation, success is critically dependent on the chosen set of features. Methodology We present an algorithmic framework (EFFECT) for automated detection of functional signals in biological sequences. We focus here on classification problems involving DNA sequences which state-of-the-art work in machine learning shows to be challenging and involve complex combinations of local and distal features. EFFECT uses a two-stage process to first construct a set of candidate sequence-based features and then select a most effective subset for the classification task at hand. Both stages make heavy use of evolutionary algorithms to efficiently guide the search towards informative features capable of discriminating between sequences that contain a particular functional signal and those that do not. Results To demonstrate its generality, EFFECT is applied to three separate problems of importance in DNA research: the recognition of hypersensitive sites, splice sites, and ALU sites. Comparisons with state-of-the-art algorithms show that the framework is both general and powerful. In addition, a detailed analysis of the constructed features shows that they contain valuable biological information about DNA architecture, allowing biologists and other researchers to directly inspect the features and potentially use the insights obtained to assist wet-laboratory studies on retainment or modification of a specific signal. Code, documentation, and all data for the applications presented here are provided for the community at http://www.cs.gmu.edu/~ashehu/?q=OurTools.

Using evolutionary computation to improve SVM classification

Support vector machines (SVMs) are now one of the most popular machine learning techniques for solving difficult classification problems. Their effectiveness depends on two critical design decisions: 1) mapping a decision problem into an n-dimensional feature space, and 2) choosing a kernel function that maps the n-dimensional feature space into a higher dimensional and more effective classification space. The choice of kernel functions is generally limited to a small set of well-studied candidates. However, the choice of a feature set is much more open-ended without much design guidance. In fact, many SVMs are designed with standard generic feature space mappings embedded a priori. In this paper we describe a procedure for using an evolutionary algorithm to design more compact non-standard feature mappings that, for a fixed kernel function, significantly improves the classification accuracy of the constructed SVM.

Selecting predictive features for recognition of hypersensitive sites of regulatory genomic sequences with an evolutionary algorithm

This paper proposes a method to improve the recognition of regulatory genomic sequences. Annotating sequences that regulate gene transcription is an emerging challenge in genomics research. Identifying regulatory sequences promises to reveal underlying reasons for phenotypic differences among cells and for diseases associated with pathologies in protein expression. Computational approaches have been limited by the scarcity of experimentally-known features specific to regulatory sequences. High-throughput experimental technology is finally revealing a wealth of hypersensitive (HS) sequences that are reliable markers of regulatory sequences and currently the focus of classification methods. The contribution of this paper is a novel method that combines evolutionary computation and SVM classification to improve the recognition of HS sequences. Based on experimental evidence that HS regions employ sequence features to interact with enzymes, the method seeks motifs to discriminate between HS and non-HS sequences. An evolutionary algorithm (EA) searches the space of sequences of different lengths to obtain such motifs. Experiments reveal that these motifs improve recognition of HS sequences by more than 10% compared to state-of-the-art classification methods. Analysis of these motifs reveals interesting insight into features employed by regulatory sequences to interact with DNA-binding enzymes.

A spatial EA framework for parallelizing machine learning methods

The scalability of machine learning (ML) algorithms has become increasingly important due to the ever increasing size of datasets and increasing complexity of the models induced. Standard approaches for dealing with this issue generally involve developing parallel and distributed versions of the ML algorithms and/or reducing the dataset sizes via sampling techniques. In this paper we describe an alternative approach that combines features of spatially-structured evolutionary algorithms (SSEAs) with the well-known machine learning techniques of ensemble learning and boosting. The result is a powerful and robust framework for parallelizing ML methods in a way that does not require changes to the ML methods. We first describe the framework and illustrate its behavior on a simple synthetic problem, and then evaluate its scalability and robustness using several different ML methods on a set of benchmark problems from the UC Irvine ML database.

A TWO-STAGE EVOLUTIONARY APPROACH FOR EFFECTIVE CLASSIFICATION OF HYPERSENSITIVE DNA SEQUENCES

Hypersensitive (HS) sites in genomic sequences are reliable markers of DNA regulatory regions that control gene expression. Annotation of regulatory regions is important in understanding phenotypical differences among cells and diseases linked to pathologies in protein expression. Several computational techniques are devoted to mapping out regulatory regions in DNA by initially identifying HS sequences. Statistical learning techniques like Support Vector Machines (SVM), for instance, are employed to classify DNA sequences as HS or non-HS. This paper proposes a method to automate the basic steps in designing an SVM that improves the accuracy of such classification. The method proceeds in two stages and makes use of evolutionary algorithms. An evolutionary algorithm first designs optimal sequence motifs to associate explicit discriminating feature vectors with input DNA sequences. A second evolutionary algorithm then designs SVM kernel functions and parameters that optimally separate the HS and non-HS classes. Results show that this two-stage method significantly improves SVM classification accuracy. The method promises to be generally useful in automating the analysis of biological sequences, and we post its source code on our website.

Feature and Kernel Evolution for Recognition of Hypersensitive Sites in DNA Sequences

The annotation of DNA regions that regulate gene transcription is the first step towards understanding phenotypical differences among cells and many diseases. Hypersensitive (HS) sites are reliable markers of regulatory regions. Mapping HS sites is the focus of many statistical learning techniques that employ Support Vector Machines (SVM) to classify a DNA sequence as HS or non-HS. The contribution of this paper is a novel methodology inspired by biological evolution to automate the basic steps in SVM and improve classification accuracy. First, an evolutionary algorithm designs optimal sequence motifs used to associate feature vectors with the input sequences. Second, a genetic programming algorithm designs optimal kernel functions that map the feature vectors into a high-dimensional space where the vectors can be optimally separated into the HS and non-HS classes. Results show that the employment of evolutionary computation techniques improves classification accuracy and promises to automate the analysis of biological sequences.

Deep learning improves antimicrobial peptide recognition

Abstract Motivation Bacterial resistance to antibiotics is a growing concern. Antimicrobial peptides (AMPs), natural components of innate immunity, are popular targets for developing new drugs. Machine learning methods are now commonly adopted by wet-laboratory researchers to screen for promising candidates. Results In this work, we utilize deep learning to recognize antimicrobial activity. We propose a neural network model with convolutional and recurrent layers that leverage primary sequence composition. Results show that the proposed model outperforms state-of-the-art classification models on a comprehensive dataset. By utilizing the embedding weights, we also present a reduced-alphabet representation and show that reasonable AMP recognition can be maintained using nine amino acid types. Availability and implementation Models and datasets are made freely available through the Antimicrobial Peptide Scanner vr.2 web server at www.ampscanner.com. Supplementary information Supplementary data are available at Bioinformatics online.

Improving Recognition of Antimicrobial Peptides and Target Selectivity through Machine Learning and Genetic Programming

Growing bacterial resistance to antibiotics is spurring research on utilizing naturally-occurring antimicrobial peptides (AMPs) as templates for novel drug design. While experimentalists mainly focus on systematic point mutations to measure the effect on antibacterial activity, the computational community seeks to understand what determines such activity in a machine learning setting. The latter seeks to identify the biological signals or features that govern activity. In this paper, we advance research in this direction through a novel method that constructs and selects complex sequence-based features which capture information about distal patterns within a peptide. Comparative analysis with state-of-the-art methods in AMP recognition reveals our method is not only among the top performers, but it also provides transparent summarizations of antibacterial activity at the sequence level. Moreover, this paper demonstrates for the first time the capability not only to recognize that a peptide is an AMP or not but also to predict its target selectivity based on models of activity against only Gram-positive, only Gram-negative, or both types of bacteria. The work described in this paper is a step forward in computational research seeking to facilitate AMP design or modification in the wet laboratory.

Boosted mean shift clustering

Mean shift is a nonparametric clustering technique that does not require the number of clusters in input and can find clusters of arbitrary shapes. While appealing, the performance of the mean shift algorithm is sensitive to the selection of the bandwidth, and can fail to capture the correct clustering structure when multiple modes exist in one cluster. DBSCAN is an efficient density based clustering algorithm, but it is also sensitive to its parameters and typically merges overlapping clusters. In this paper we propose Boosted Mean Shift Clustering (BMSC) to address these issues. BMSC partitions the data across a grid and applies mean shift locally on the cells of the grid, each providing a number of intermediate modes (iModes). A mode-boosting technique is proposed to select points in denser regions iteratively, and DBSCAN is utilized to partition the obtained iModes iteratively. Our proposed BMSC can overcome the limitations of mean shift and DBSCAN, while preserving their desirable properties. Complexity analysis shows its potential to deal with large-scale data and extensive experimental results on both synthetic and real benchmark data demonstrate its effectiveness and robustness to parameter settings.