We present a method utilizing Healthcare Cost and Utilization Project (HCUP) dataset for predicting disease risk of individuals based on their medical diagnosis history. The presented methodology may be incorporated in a variety of applications such as risk management, tailored health communication and decision support systems in healthcare. We employed the National Inpatient Sample (NIS) data, which is publicly available through Healthcare Cost and Utilization Project (HCUP), to train random forest classifiers for disease prediction. Since the HCUP data is highly imbalanced, we employed an ensemble learning approach based on repeated random sub-sampling. This technique divides the training data into multiple sub-samples, while ensuring that each sub-sample is fully balanced. We compared the performance of support vector machine (SVM), bagging, boosting and RF to predict the risk of eight chronic diseases. We predicted eight disease categories. Overall, the RF ensemble learning method outperformed SVM, bagging and boosting in terms of the area under the receiver operating characteristic (ROC) curve (AUC). In addition, RF has the advantage of computing the importance of each variable in the classification process. In combining repeated random sub-sampling with RF, we were able to overcome the class imbalance problem and achieve promising results. Using the national HCUP data set, we predicted eight disease categories with an average AUC of 88.79%.

https://link.springer.com/content/pdf/10.1186%2F1472-6947-11-51.pdf

Predicting disease risks from highly imbalanced data using random forest

Innovations in sequencing technologies have allowed biologists to make incredible advances in understanding biological systems. As experience grows, researchers increasingly recognize that analyzing the wealth of data provided by these new sequencing platforms requires careful attention to detail for robust results. Thus far, much of the scientific Communit's focus for use in bacterial genomics has been on evaluating genome assembly algorithms and rigorously validating assembly program performance. Missing, however, is a focus on critical evaluation of variant callers for these genomes. Variant calling is essential for comparative genomics as it yields insights into nucleotide-level organismal differences. Variant calling is a multistep process with a host of potential error sources that may lead to incorrect variant calls. Identifying and resolving these incorrect calls is critical for bacterial genomics to advance. The goal of this review is to provide guidance on validating algorithms and pipelines used in variant calling for bacterial genomics. First, we will provide an overview of the variant calling procedures and the potential sources of error associated with the methods. We will then identify appropriate datasets for use in evaluating algorithms and describe statistical methods for evaluating algorithm performance. As variant calling moves from basic research to the applied setting, standardized methods for performance evaluation and reporting are required; it is our hope that this review provides the groundwork for the development of these standards.

/pdf/best-practices-for-evaluating-single-nucleotide-variant-1r1i2aj34r.pdf

Best practices for evaluating single nucleotide variant calling methods for microbial genomics.

Reviewing ensemble classification methods in breast cancer.

Motivation Oxford Nanopore sequencing is a rapidly developed sequencing technology in recent years. To keep pace with the explosion of the downstream data analytical tools, a versatile Nanopore sequencing simulator is needed to complement the experimental data as well as to benchmark those newly developed tools. However, all the currently available simulators are based on simple statistics of the produced reads, which have difficulty in capturing the complex nature of the Nanopore sequencing procedure, the main task of which is the generation of raw electrical current signals. Results Here we propose a deep learning based simulator, DeepSimulator, to mimic the entire pipeline of Nanopore sequencing. Starting from a given reference genome or assembled contigs, we simulate the electrical current signals by a context-dependent deep learning model, followed by a base-calling procedure to yield simulated reads. This workflow mimics the sequencing procedure more naturally. The thorough experiments performed across four species show that the signals generated by our context-dependent model are more similar to the experimentally obtained signals than the ones generated by the official context-independent pore model. In terms of the simulated reads, we provide a parameter interface to users so that they can obtain the reads with different accuracies ranging from 83 to 97%. The reads generated by the default parameter have almost the same properties as the real data. Two case studies demonstrate the application of DeepSimulator to benefit the development of tools in de novo assembly and in low coverage SNP detection. Availability and implementation The software can be accessed freely at: https://github.com/lykaust15/DeepSimulator. Supplementary information Supplementary data are available at Bioinformatics online.

/pdf/deepsimulator-a-deep-simulator-for-nanopore-sequencing-4782ubq0j1.pdf

DeepSimulator: a deep simulator for Nanopore sequencing.

A new approach to solve the power flow problem based on graphic processing units is presented in this paper. A Newton method is implemented to solve the set of nonlinear equations of the power flow formulation. A parallel kernel for the biconjugate gradient method allows solving the voltage corrections on a graphic processing card. While the evaluation of the Jacobian matrix is carried out on the CPU of the host PC, the most demanding task of the iterative biconjugate gradient method for the solution of a linear system is performed on the GPU. Furthermore, a row-indexed sparse storage scheme implemented in the host PC and on the device graphic card allows providing efficient solutions in terms of storage and computational effort. A test case based on the IEEE-118 node system is solved to show the accuracy and efficiency of power flow solutions computed on the graphic processing unit.

Parallel power flow solutions using a biconjugate gradient algorithm and a Newton method: A GPU-based approach

Motivation: The identification of short insertions and deletions (indels) and single nucleotide polymorphisms (SNPs) from Ion Torrent and 454 reads is a challenging problem, essentially because these techniques are prone to sequence erroneously at homopolymers and can, therefore, raise indels in reads. Most of the existing mapping programs do not model homopolymer errors when aligning reads against the reference. The resulting alignments will then contain various kinds of mismatches and indels that confound the accurate determination of variant loci and alleles. Results: To address these challenges, we realign reads against the reference using our previously proposed hidden Markov model that models homopolymer errors and then merges these pairwise alignments into a weighted alignment graph. Based on our weighted alignment graph and hidden Markov model, we develop a method called PyroHMMvar, which can simultaneously detect short indels and SNPs, as demonstrated in human resequencing data. Specifically, by applying our methods to simulated diploid datasets, we demonstrate that PyroHMMvar produces more accurate results than state-of-the-art methods, such as Samtools and GATK, and is less sensitive to mapping parameter settings than the other methods. We also apply PyroHMMvar to analyze one human whole genome resequencing dataset, and the results confirm that PyroHMMvar pre

/pdf/pyrohmmvar-a-sensitive-and-accurate-method-to-call-short-3zi1nvfjaw.pdf

PyroHMMvar: a sensitive and accurate method to call short indels and SNPs for Ion Torrent and 454 data.

The combined use of gene expression profiles and protein-protein interaction (PPI) networks has recently shed light on breast cancer research by selecting a small number of subnetworks as disease markers and then using them for the classification of metastasis Based on previously identified subnetwork markers, we compare three ensemble learning approaches (AdaBoost, LogitBoost and random forest) with two widely used classifiers (logistic regression and support vector machine) in the classification of breast cancer metastasis In leave-one-out cross-validation experiments on two breast cancer data sets, the ensemble learning methods can lead logis-tic regression and support vector machine by 224% and 48% respectively in terms of the classification accuracy In cross data set validation experiments, the ensemble learning methods also demonstrate superior reproducibility over the other two methods With these results, we infer that the ensemble learn-ing approaches with subnetwork markers might be more suit-able in handling the classification problem of breast cancer metastasis, and we recommend the use of these approaches in similar classification problems

/pdf/a-comparative-study-of-ensemble-learning-approaches-in-the-2ppec4pl7h.pdf

A Comparative Study of Ensemble Learning Approaches in the Classification of Breast Cancer Metastasis

Both 454 and Ion Torrent sequencers are capable of producing large amounts of long high-quality sequencing reads. However, as both methods sequence homopolymers in one cycle, they both suffer from homopolymer uncertainty and incorporation asynchronization. In mapping, such sequencing errors could shift alignments around homopolymers and thus induce incorrect mismatches, which have become a critical barrier against the accurate detection of single nucleotide polymorphisms (SNPs). In this article, we propose a hidden Markov model (HMM) to statistically and explicitly formulate homopolymer sequencing errors by the overcall, undercall, insertion and deletion. We use a hierarchical model to describe the sequencing and base-calling processes, and we estimate parameters of the HMM from resequencing data by an expectation-maximization algorithm. Based on the HMM, we develop a realignment-based SNP-calling program, termed PyroHMMsnp, which realigns read sequences around homopolymers according to the error model and then infers the underlying genotype by using a Bayesian approach. Simulation experiments show that the performance of PyroHMMsnp is exceptional across various sequencing coverages in terms of sensitivity, specificity and F1 measure, compared with other tools. Analysis of the human resequencing data shows that PyroHMMsnp predicts 12.9% more SNPs than Samtools while achieving a higher specificity. (http://code.google.com/p/pyrohmmsnp/).

PyroHMMsnp: an SNP caller for Ion Torrent and 454 sequencing data

Recent advances in highly parallel, multithreaded, manycore Graphics Processing Units (GPUs) have been enabling massive parallel implementations of many applications in bioinformatics. In this paper, we describe a parallel implementation of genome-wide association studies (GWAS) using Compute Unified Device Architecture (CUDA). Using a single NVIDIA GTX 280 graphics card, we achieve speedups of about 15 times over Intel Xeon E5420. We also implement a highly scalable, massive parallel, GWAS system using the Message Passing Interface (MPI) and show that a single GTX 280 can have similar performance as a 16-node cluster. We further apply the GPU program to two real genome-wide case-control data sets. The results show that the GPU program is 17.7 times as fast as the CPU version for an Age-related Macular Degeneration (AMD) data set and 25.7 times as fast as the CPU version for a Parkinson’s disease data set.

Accelerating Genome-Wide Association Studies Using CUDA Compatible Graphics Processing Units

Identifying different types of cells in scRNA-seq data is a critical task in single-cell data analysis. In this paper, we propose a method called ProgClust for the decomposition of cell populations and detection of rare cells. ProgClust represents the single-cell data with clustering trees where a progressive searching method is designed to select cell population-specific genes and cluster cells. The obtained trees reveal the structure of both abundant cell populations and rare cell populations. Additionally, it can automatically determine the number of clusters. Experimental results show that ProgClust outperforms the baseline method and is capable of accurately identifying both common and rare cells. Moreover, when applied to real unlabeled data, it reveals potential cell subpopulations which provides clues for further exploration. In summary, ProgClust shows potential in identifying subpopulations of complex single-cell data.

/pdf/progclust-a-progressive-clustering-method-to-identify-cell-17t5jsmm.pdf

Feng Zeng

Papers

PyroHMMvar: a sensitive and accurate method to call short indels and SNPs for Ion Torrent and 454 data.

A Comparative Study of Ensemble Learning Approaches in the Classification of Breast Cancer Metastasis

PyroHMMsnp: an SNP caller for Ion Torrent and 454 sequencing data

Accelerating Genome-Wide Association Studies Using CUDA Compatible Graphics Processing Units

ProgClust: A progressive clustering method to identify cell populations