Deep learning for DNase I hypersensitive sites identification

Background

The DNase I hypersensitive sites (DHSs) are associated with the cis-regulatory DNA elements. An efficient method of identifying DHSs can enhance the understanding on the accessibility of chromatin. Despite a multitude of resources available on line including experimental datasets and computational tools, the complex language of DHSs remains incompletely understood.

Methods

Here, we address this challenge using an approach based on a state-of-the-art machine learning method. We present a novel convolutional neural network (CNN) which combined Inception like networks with a gating mechanism for the response of multiple patterns and longterm association in DNA sequences to predict multi-scale DHSs in Arabidopsis, rice and Homo sapiens.

Results

Our method obtains 0.961 area under curve (AUC) on Arabidopsis, 0.969 AUC on rice and 0.918 AUC on Homo sapiens.

Conclusions

Our method provides an efficient and accurate way to identify multi-scale DHSs sequences by deep learning.

     

An effective evolutionary algorithm for protein folding on 3D FCC HP model by lattice rotation and generalized move sets

Background

Proteins are essential biological molecules which play vital roles in nearly all biological processes. It is the tertiary structure of a protein that determines its functions. Therefore the prediction of a protein's tertiary structure based on its primary amino acid sequence has long been the most important and challenging subject in biochemistry, molecular biology and biophysics. In the past, the HP lattice model was one of the ab initio methods that many researchers used to forecast the protein structure. Although these kinds of simplified methods could not achieve high resolution, they provided a macrocosm-optimized protein structure. The model has been employed to investigate general principles of protein folding, and plays an important role in the prediction of protein structures.

Methods

In this paper, we present an improved evolutionary algorithm for the protein folding problem. We study the problem on the 3D FCC lattice HP model which has been widely used in previous research. Our focus is to develop evolutionary algorithms (EA) which are robust, easy to implement and can handle various energy functions. We propose to combine three different local search methods, including lattice rotation for crossover, K-site move for mutation, and generalized pull move; these form our key components to improve previous EA-based approaches.

Results

We have carried out experiments over several data sets which were used in previous research. The results of the experiments show that our approach is able to find optimal conformations which were not found by previous EA-based approaches.

Conclusions

We have investigated the geometric properties of the 3D FCC lattice and developed several local search techniques to improve traditional EA-based approaches to the protein folding problem. It is known that EA-based approaches are robust and can handle arbitrary energy functions. Our results further show that by extensive development of local searches, EA can also be very effective for finding optimal conformations on the 3D FCC HP model. Furthermore, the local searches developed in this paper can be integrated with other approaches such as the Monte Carlo and Tabu searches to improve their performance.

     

An effective hybrid of hill climbing and genetic algorithm for 2D triangular protein structure prediction

Background

Proteins play fundamental and crucial roles in nearly all biological processes, such as, enzymatic catalysis, signaling transduction, DNA and RNA synthesis, and embryonic development. It has been a long-standing goal in molecular biology to predict the tertiary structure of a protein from its primary amino acid sequence. From visual comparison, it was found that a 2D triangular lattice model can give a better structure modeling and prediction for proteins with short primary amino acid sequences.

Methods

This paper proposes a hybrid of hill-climbing and genetic algorithm (HHGA) based on elite-based reproduction strategy for protein structure prediction on the 2D triangular lattice.

Results

The simulation results show that the proposed HHGA can successfully deal with the protein structure prediction problems. Specifically, HHGA significantly outperforms conventional genetic algorithms and is comparable to the state-of-the-art method in terms of free energy.

Conclusions

Thanks to the enhancement of local search on the global search, the proposed HHGA achieves promising results on the 2D triangular protein structure prediction problem. The satisfactory simulation results demonstrate the effectiveness of the proposed HHGA and the utility of the 2D triangular lattice model for protein structure prediction.

     

Predicting adverse drug reactions through interpretable deep learning framework

Background

Adverse drug reactions (ADRs) are unintended and harmful reactions caused by normal uses of drugs. Predicting and preventing ADRs in the early stage of the drug development pipeline can help to enhance drug safety and reduce financial costs.

Methods

In this paper, we developed machine learning models including a deep learning framework which can simultaneously predict ADRs and identify the molecular substructures associated with those ADRs without defining the substructures a-priori.

Results

We evaluated the performance of our model with ten different state-of-the-art fingerprint models and found that neural fingerprints from the deep learning model outperformed all other methods in predicting ADRs. Via feature analysis on drug structures, we identified important molecular substructures that are associated with specific ADRs and assessed their associations via statistical analysis.

Conclusions

The deep learning model with feature analysis, substructure identification, and statistical assessment provides a promising solution for identifying risky components within molecular structures and can potentially help to improve drug safety evaluation.

     

Comparison of named entity recognition methodologies in biomedical documents

Background

Biomedical named entity recognition (Bio-NER) is a fundamental task in handling biomedical text terms, such as RNA, protein, cell type, cell line, and DNA. Bio-NER is one of the most elementary and core tasks in biomedical knowledge discovery from texts. The system described here is developed by using the BioNLP/NLPBA 2004 shared task. Experiments are conducted on a training and evaluation set provided by the task organizers.

Results

Our results show that, compared with a baseline having a 70.09% F1 score, the RNN Jordan- and Elman-type algorithms have F1 scores of approximately 60.53% and 58.80%, respectively. When we use CRF as a machine learning algorithm, CCA, GloVe, and Word2Vec have F1 scores of 72.73%, 72.74%, and 72.82%, respectively.

Conclusions

By using the word embedding constructed through the unsupervised learning, the time and cost required to construct the learning data can be saved.

     

Using multitask classification methods to investigate the kinase-specific phosphorylation sites

Background

Identification of phosphorylation sites by computational methods is becoming increasingly important because it reduces labor-intensive and costly experiments and can improve our understanding of the common properties and underlying mechanisms of protein phosphorylation.

Methods

A multitask learning framework for learning four kinase families simultaneously, instead of studying each kinase family of phosphorylation sites separately, is presented in the study. The framework includes two multitask classification methods: the Multi-Task Least Squares Support Vector Machines (MTLS-SVMs) and the Multi-Task Feature Selection (MT-Feat3).

Results

Using the multitask learning framework, we successfully identify 18 common features shared by four kinase families of phosphorylation sites. The reliability of selected features is demonstrated by the consistent performance in two multi-task learning methods.

Conclusions

The selected features can be used to build efficient multitask classifiers with good performance, suggesting they are important to protein phosphorylation across 4 kinase families.

     

Extracting DNA words based on the sequence features: non-uniform distribution and integrity

Background

DNA sequence can be viewed as an unknown language with words as its functional units. Given that most sequence alignment algorithms such as the motif discovery algorithms depend on the quality of background information about sequences, it is necessary to develop an ab initio algorithm for extracting the “words” based only on the DNA sequences.

Methods

We considered that non-uniform distribution and integrity were two important features of a word, based on which we developed an ab initio algorithm to extract “DNA words” that have potential functional meaning. A Kolmogorov-Smirnov test was used for consistency test of uniform distribution of DNA sequences, and the integrity was judged by the sequence and position alignment. Two random base sequences were adopted as negative control, and an English book was used as positive control to verify our algorithm. We applied our algorithm to the genomes of Saccharomyces cerevisiae and 10 strains of Escherichia coli to show the utility of the methods.

Results

The results provide strong evidences that the algorithm is a promising tool for ab initio building a DNA dictionary.

Conclusions

Our method provides a fast way for large scale screening of important DNA elements and offers potential insights into the understanding of a genome.

     

Hot spot prediction in protein-protein interactions by an ensemble system

Background

Hot spot residues are functional sites in protein interaction interfaces. The identification of hot spot residues is time-consuming and laborious using experimental methods. In order to address the issue, many computational methods have been developed to predict hot spot residues. Moreover, most prediction methods are based on structural features, sequence characteristics, and/or other protein features.

Results

This paper proposed an ensemble learning method to predict hot spot residues that only uses sequence features and the relative accessible surface area of amino acid sequences. In this work, a novel feature selection technique was developed, an auto-correlation function combined with a sliding window technique was applied to obtain the characteristics of amino acid residues in protein sequence, and an ensemble classifier with SVM and KNN base classifiers was built to achieve the best classification performance.

Conclusion

The experimental results showed that our model yields the highest F1 score of 0.92 and an MCC value of 0.87 on ASEdb dataset. Compared with other machine learning methods, our model achieves a big improvement in hot spot prediction.

Availability

http://deeplearner.ahu.edu.cn/web/HotspotEL.htm.

     

Intrinsic disorder in putative protein sequences

Background

Intrinsically disordered proteins (IDPs) and regions (IDRs) perform a variety of crucial biological functions despite lacking stable tertiary structure under physiological conditions in vitro. State-of-the-art sequence-based predictors of intrinsic disorder are achieving per-residue accuracies over 80%. In a genome-wide study of intrinsic disorder in human genome we observed a big difference in predicted disorder content between confirmed and putative human proteins. We investigated a hypothesis that this discrepancy is not correct, and that it is due to incorrectly annotated parts of the putative protein sequences that exhibit some similarities to confirmed IDRs, which lead to high predicted disorder content.

Methods

To test this hypothesis we trained a predictor to discriminate sequences of real proteins from synthetic sequences that mimic errors of gene finding algorithms. We developed a procedure to create synthetic peptide sequences by translation of non-coding regions of genomic sequences and translation of coding regions with incorrect codon alignment.

Results

Application of the developed predictor to putative human protein sequences showed that they contain a substantial fraction of incorrectly assigned regions. These regions are predicted to have higher levels of disorder content than correctly assigned regions. This partially, albeit not completely, explains the observed discrepancy in predicted disorder content between confirmed and putative human proteins.

Conclusions

Our findings provide the first evidence that current practice of predicting disorder content in putative sequences should be reconsidered, as such estimates may be biased.

     

Predicting enhancers with deep convolutional neural networks

Background

With the rapid development of deep sequencing techniques in the recent years, enhancers have been systematically identified in such projects as FANTOM and ENCODE, forming genome-wide landscapes in a series of human cell lines. Nevertheless, experimental approaches are still costly and time consuming for large scale identification of enhancers across a variety of tissues under different disease status, making computational identification of enhancers indispensable.

Results

To facilitate the identification of enhancers, we propose a computational framework, named DeepEnhancer, to distinguish enhancers from background genomic sequences. Our method purely relies on DNA sequences to predict enhancers in an end-to-end manner by using a deep convolutional neural network (CNN). We train our deep learning model on permissive enhancers and then adopt a transfer learning strategy to fine-tune the model on enhancers specific to a cell line. Results demonstrate the effectiveness and efficiency of our method in the classification of enhancers against random sequences, exhibiting advantages of deep learning over traditional sequence-based classifiers. We then construct a variety of neural networks with different architectures and show the usefulness of such techniques as max-pooling and batch normalization in our method. To gain the interpretability of our approach, we further visualize convolutional kernels as sequence logos and successfully identify similar motifs in the JASPAR database.

Conclusions

DeepEnhancer enables the identification of novel enhancers using only DNA sequences via a highly accurate deep learning model. The proposed computational framework can also be applied to similar problems, thereby prompting the use of machine learning methods in life sciences.

     

Automatic infection detection based on electronic medical records

Background

Making accurate patient care decision, as early as possible, is a constant challenge, especially for physicians in the emergency department. The increasing volumes of electronic medical records (EMRs) open new horizons for automatic diagnosis. In this paper, we propose to use machine learning approaches for automatic infection detection based on EMRs. Five categories of information are utilized for prediction, including personal information, admission note, vital signs, diagnose test results and medical image diagnose.

Results

Experimental results on a newly constructed EMRs dataset from emergency department show that machine learning models can achieve a decent performance for infection detection with area under the receiver operator characteristic curve (AUC) of 0.88. Out of all the five types of information, admission note in text form makes the most contribution with the AUC of 0.87.

Conclusions

This study provides a state-of-the-art EMRs processing system to automatically make medical decisions. It extracts five types of features associated with infection and achieves a decent performance on automatic infection detection based on machine learning models.

     

Improved de novo peptide sequencing using LC retention time information

Background

Liquid chromatography combined with tandem mass spectrometry is an important tool in proteomics for peptide identification. Liquid chromatography temporally separates the peptides in a sample. The peptides that elute one after another are analyzed via tandem mass spectrometry by measuring the mass-to-charge ratio of a peptide and its fragments. De novo peptide sequencing is the problem of reconstructing the amino acid sequences of a peptide from this measurement data. Past de novo sequencing algorithms solely consider the mass spectrum of the fragments for reconstructing a sequence.

Results

We propose to additionally exploit the information obtained from liquid chromatography. We study the problem of computing a sequence that is not only in accordance with the experimental mass spectrum, but also with the chromatographic retention time. We consider three models for predicting the retention time and develop algorithms for de novo sequencing for each model.

Conclusions

Based on an evaluation for two prediction models on experimental data from synthesized peptides we conclude that the identification rates are improved by exploiting the chromatographic information. In our evaluation, we compare our algorithms using the retention time information with algorithms using the same scoring model, but not the retention time.

     

An interpretable boosting model to predict side effects of analgesics for osteoarthritis

Background

Osteoarthritis (OA) is the most common disease of arthritis. Analgesics are widely used in the treat of arthritis, which may increase the risk of cardiovascular diseases by 20% to 50% overall.There are few studies on the side effects of OA medication, especially the risk prediction models on side effects of analgesics. In addition, most prediction models do not provide clinically useful interpretable rules to explain the reasoning process behind their predictions. In order to assist OA patients, we use the eXtreme Gradient Boosting (XGBoost) method to balance the accuracy and interpretability of the prediction model.

Results

In this study we used the XGBoost model as a classifier, which is a supervised machine learning method and can predict side effects of analgesics for OA patients and identify high-risk features (RFs) of cardiovascular diseases caused by analgesics. The Electronic Medical Records (EMRs), which were derived from public knee OA studies, were used to train the model. The performance of the XGBoost model is superior to four well-known machine learning algorithms and identifies the risk features from the biomedical literature. In addition the model can provide decision support for using analgesics in OA patients.

Conclusion

Compared with other machine learning methods, we used XGBoost method to predict side effects of analgesics for OA patients from EMRs, and selected the individual informative RFs. The model has good predictability and interpretability, this is valuable for both medical researchers and patients.

     

Detection of potential new biomarkers of atherosclerosis by probe electrospray ionization mass spectrometry

Introduction

Atherosclerotic diseases are the leading cause of death worldwide. Biomarkers of atherosclerosis are required to monitor and prevent disease progression. While mass spectrometry is a promising technique to search for such biomarkers, its clinical application is hampered by the laborious processes for sample preparation and analysis.

Methods

We developed a rapid method to detect plasma metabolites by probe electrospray ionization mass spectrometry (PESI-MS), which employs an ambient ionization technique enabling atmospheric pressure rapid mass spectrometry. To create an automatic diagnosis system of atherosclerotic disorders, we applied machine learning techniques to the obtained spectra.

Results

Using our system, we successfully discriminated between rabbits with and without dyslipidemia. The causes of dyslipidemia (genetic lipoprotein receptor deficiency or dietary cholesterol overload) were also distinguishable by this method. Furthermore, after induction of atherosclerosis in rabbits with a cholesterol-rich diet, we were able to detect dynamic changes in plasma metabolites. The major metabolites detected by PESI-MS included cholesterol sulfate and a phospholipid (PE18:0/20:4), which are promising new biomarkers of atherosclerosis.

Conclusion

We developed a remarkably fast and easy method to detect potential new biomarkers of atherosclerosis in plasma using PESI-MS.

     

PROPER: Performance visualization for optimizing and comparing ranking classifiers in MATLAB

Background

One of the recent challenges of computational biology is development of new algorithms, tools and software to facilitate predictive modeling of big data generated by high-throughput technologies in biomedical research.

Results

To meet these demands we developed PROPER - a package for visual evaluation of ranking classifiers for biological big data mining studies in the MATLAB environment.

Conclusion

PROPER is an efficient tool for optimization and comparison of ranking classifiers, providing over 20 different two- and three-dimensional performance curves.

     

Optimization of fecal sample preparation for untargeted LC-HRMS based metabolomics

Introduction

Collecting feces is easy. It offers direct outcome to endogenous and microbial metabolites.

Objectives

In a context of lack of consensus about fecal sample preparation, especially in animal species, we developed a robust protocol allowing untargeted LC-HRMS fingerprinting.

Methods

The conditions of extraction (quantity, preparation, solvents, dilutions) were investigated in bovine feces.

Results

A rapid and simple protocol involving feces extraction with methanol (1/3, M/V) followed by centrifugation and a step filtration (10 kDa) was developed.

Conclusion

The workflow generated repeatable and informative fingerprints for robust metabolome characterization.

     

RiboTagger: fast and unbiased 16S/18S profiling using whole community shotgun metagenomic or metatranscriptome surveys

Background

Taxonomic profiling of microbial communities is often performed using small subunit ribosomal RNA (SSU) amplicon sequencing (16S or 18S), while environmental shotgun sequencing is often focused on functional analysis. Large shotgun datasets contain a significant number of SSU sequences and these can be exploited to perform an unbiased SSU--based taxonomic analysis.

Results

Here we present a new program called RiboTagger that identifies and extracts taxonomically informative ribotags located in a specified variable region of the SSU gene in a high-throughput fashion.

Conclusions

RiboTagger permits fast recovery of SSU-RNA sequences from shotgun nucleic acid surveys of complex microbial communities. The program targets all three domains of life, exhibits high sensitivity and specificity and is substantially faster than comparable programs.

     

Learning from positive examples when the negative class is undetermined- microRNA gene identification

Background

The application of machine learning to classification problems that depend only on positive examples is gaining attention in the computational biology community. We and others have described the use of two-class machine learning to identify novel miRNAs. These methods require the generation of an artificial negative class. However, designation of the negative class can be problematic and if it is not properly done can affect the performance of the classifier dramatically and/or yield a biased estimate of performance. We present a study using one-class machine learning for microRNA (miRNA) discovery and compare one-class to two-class approaches using naïve Bayes and Support Vector Machines. These results are compared to published two-class miRNA prediction approaches. We also examine the ability of the one-class and two-class techniques to identify miRNAs in newly sequenced species.

Results

Of all methods tested, we found that 2-class naive Bayes and Support Vector Machines gave the best accuracy using our selected features and optimally chosen negative examples. One class methods showed average accuracies of 70–80% versus 90% for the two 2-class methods on the same feature sets. However, some one-class methods outperform some recently published two-class approaches with different selected features. Using the EBV genome as and external validation of the method we found one-class machine learning to work as well as or better than a two-class approach in identifying true miRNAs as well as predicting new miRNAs.

Conclusion

One and two class methods can both give useful classification accuracies when the negative class is well characterized. The advantage of one class methods is that it eliminates guessing at the optimal features for the negative class when they are not well defined. In these cases one-class methods can be superior to two-class methods when the features which are chosen as representative of that positive class are well defined.

Availability

The OneClassmiRNA program is available at: [1]

     

Automatic recognition of topic-classified relations between prostate cancer and genes using MEDLINE abstracts

Background

Automatic recognition of relations between a specific disease term and its relevant genes or protein terms is an important practice of bioinformatics. Considering the utility of the results of this approach, we identified prostate cancer and gene terms with the ID tags of public biomedical databases. Moreover, considering that genetics experts will use our results, we classified them based on six topics that can be used to analyze the type of prostate cancers, genes, and their relations.

Methods

We developed a maximum entropy-based named entity recognizer and a relation recognizer and applied them to a corpus-based approach. We collected prostate cancer-related abstracts from MEDLINE, and constructed an annotated corpus of gene and prostate cancer relations based on six topics by biologists. We used it to train the maximum entropy-based named entity recognizer and relation recognizer.

Results

Topic-classified relation recognition achieved 92.1% precision for the relation (an increase of 11.0% from that obtained in a baseline experiment). For all topics, the precision was between 67.6 and 88.1%.

Conclusion

A series of experimental results revealed two important findings: a carefully designed relation recognition system using named entity recognition can improve the performance of relation recognition, and topic-classified relation recognition can be effectively addressed through a corpus-based approach using manual annotation and machine learning techniques.