No abstract

An important approach for efficient support vector machine (SVM) model selection is to use differentiable bounds of the leave-one-out (loo) error. Past efforts focused on finding tight bounds of loo (e.g., radius margin bounds, span bounds). However, their practical viability is still not very satisfactory. Duan, Keerthi, and Poo (2003) showed that radius margin bound gives good prediction for L2-SVM, one of the cases we look at. In this letter, through analyses about why this bound performs well for L2-SVM, we show that finding a bound whose minima are in a region with small loo values may be more important than its tightness. Based on this principle, we propose modified radius margin bounds for L1-SVM (the other case) where the original bound is applicable only to the hard-margin case. Our modification for L1-SVM achieves comparable performance to L2-SVM. To study whether L1- or L2-SVM should be used, we analyze other properties, such as their differentiability, number of support vectors, and number of free support vectors. In this aspect, L1-SVM possesses the advantage of having fewer support vectors. Their implementations are also different, so we discuss related issues in detail.

Extracting sequence information from raw images of fluorescence is the foundation underlying several high-throughput sequencing platforms. Some of the main challenges associated with this technology include reducing the error rate, assigning accurate base-specific quality scores, and reducing the cost of sequencing by increasing the throughput per run. To demonstrate how computational advancement can help to meet these challenges, a novel model-based base-calling algorithm, BayesCall, is introduced for the Illumina sequencing platform. Being founded on the tools of statistical learning, BayesCall is flexible enough to incorporate various features of the sequencing process. In particular, it can easily incorporate time-dependent parameters and model residual effects. This new approach significantly improves the accuracy over Illumina's base-caller Bustard, particularly in the later cycles of a sequencing run. For 76-cycle data on a standard viral sample, phiX174, BayesCall improves Bustard's average per-base error rate by ;51%. The probability of observing each base can be readily computed in BayesCall, and this probability can be transformed into a useful basespecific quality score with a high discrimination ability. A detailed study of BayesCall's performance is presented here.[Supplemental material is available online at

Developing accurate, scalable algorithms to improve data quality is an important computational challenge associated with recent advances in high-throughput sequencing technology. In this study, a novel error-correction algorithm, called ECHO, is introduced for correcting base-call errors in short-reads, without the need of a reference genome. Unlike most previous methods, ECHO does not require the user to specify parameters of which optimal values are typically unknown a priori. ECHO automatically sets the parameters in the assumed model and estimates error characteristics specific to each sequencing run, while maintaining a running time that is within the range of practical use. ECHO is based on a probabilistic model and is able to assign a quality score to each corrected base. Furthermore, it explicitly models heterozygosity in diploid genomes and provides a reference-free method for detecting bases that originated from heterozygous sites. On both real and simulated data, ECHO is able to improve the accuracy of previous error-correction methods by several folds to an order of magnitude, depending on the sequence coverage depth and the position in the read. The improvement is most pronounced toward the end of the read, where previous methods become noticeably less effective. Using a wholegenome yeast data set, it is demonstrated here that ECHO is capable of coping with nonuniform coverage. Also, it is shown that using ECHO to perform error correction as a preprocessing step considerably facilitates de novo assembly, particularly in the case of low-to-moderate sequence coverage depth.[Supplemental material is available for this article. ECHO is publicly available at http://uc-echo.sourceforge.net under the Berkeley Software Distribution License.]Over the past few years, next-generation sequencing (NGS) technologies have introduced a rapidly growing wave of information in biological sciences; see Metzker (2010) for a recent review of NGS platforms and their applications. Exploiting massive parallelization, NGS platforms generate high-throughput data at very low cost per base. An important computational challenge associated with this rapid technological advancement is to develop efficient algorithms to extract accurate sequence information. In comparison with traditional Sanger sequencing (Sanger et al. 1977), NGS data have shorter read lengths and higher error rates, and these characteristics create many challenges for computation, especially when a reference genome is not available. Reducing the error rate of base-calls and improving the accuracy of base-specific quality scores have important practical implications for assembly (Sundquist et al.

Our findings demonstrate that the majority of weight decrease/dehydration in both the 12- and 24-hour races occurred during the first 8 hours. Hence, to maintain body weight, fluid intake should be optimized in the first 8 hours for both 12- and 24-hour runners and in 16 to 20 hours for 24-hour marathon runners.

Actinobacteria are closely linked to human life as industrial producers of bioactive molecules and as human pathogens. Respiratory cytochrome bcc complex and cytochrome aa3 oxidase are key components of their aerobic energy metabolism. They form a supercomplex in the actinobacterial species Corynebacterium glutamicum. With comprehensive bioinformatics and phylogenetic analysis we show that genes for cyt bcc-aa3 supercomplex are characteristic for Actinobacteria (Actinobacteria and Acidimicrobiia, except the anaerobic orders Actinomycetales and Bifidobacteriales). An obligatory supercomplex is likely, due to the lack of genes encoding alternative electron transfer partners such as mono-heme cyt c. Instead, subunit QcrC of bcc complex, here classified as short di-heme cyt c, will provide the exclusive electron transfer link between the complexes as in C. glutamicum. Purified to high homogeneity, the C. glutamicum bcc-aa3 supercomplex contained all subunits and cofactors as analyzed by SDS-PAGE, BN-PAGE, absorption and EPR spectroscopy. Highly uniform supercomplex particles in electron microscopy analysis support a distinct structural composition. The supercomplex possesses a dimeric stoichiometry with a ratio of a-type, b-type and c-type hemes close to 1:1:1. Redox titrations revealed a low potential bcc complex (Em(ISP)=+160mV, Em(bL)=-291mV, Em(bH)=-163mV, Em(cc)=+100mV) fined-tuned for oxidation of menaquinol and a mixed potential aa3 oxidase (Em(CuA)=+150mV, Em(a/a3)=+143/+317mV) mediating between low and high redox potential to accomplish dioxygen reduction. The generated molecular model supports a stable assembled supercomplex with defined architecture which permits energetically efficient coupling of menaquinol oxidation and dioxygen reduction in one supramolecular entity.

Recording of cancer diagnoses and survival estimates based on these diagnosis codes in the NHI database are generally consistent with the NCR. Studies using NHI database data must pay careful attention to eligibility and record linkage; use of both sources is recommended.

Copper ions switch the oxidation of methane by soluble methane monooxygenase to particulate methane monooxygenase in Methylococcus capsulatus (Bath). Toward understanding the change in cellular metabolism related to this transcriptional and metabolic switch, we have undertaken genomic sequencing and quantitative comparative analysis of the proteome in M. capsulatus (Bath) grown under different copper-to-biomass ratios by cleavable isotope-coded affinity tag technology. Of the 682 proteins identified, the expressions of 60 proteins were stimulated by at least 2-fold by copper ions; 68 proteins were down-regulated by 2-fold or more. The 60 proteins overexpressed included the methane and carbohydrate metabolic enzymes, while the 68 proteins suppressed were mainly responsible for cellular signaling processes, indicating a role of copper ions in the expression of the genes associated with the metabolism of the organism downstream of methane oxidation. The study has also provided a complete map of the C 1 metabolism pathways in this methanotroph and clarified the interrelationships between them.Methanotrophs are a unique group of Gram-negative bacteria that grow aerobically on methane and utilize methane as the sole source of carbon and energy. There has been considerable interest in methanotrophs over the past 30 years, since they can be used to produce single-cell bulk chemicals such as propylene oxide. The ability of these bacteria to co-oxidize a wide range of alkanes, alkenes, and substituted aliphatic compounds has been exploited in bioremediation processes, for example, in the degradation of key pollutants such as trichloroethylene in soil and groundwater. Methanotrophs also play an important role in the global methane cycle (1, 2).Methanotrophs are classified into three distinct types. Type I methanotrophs show disk-like intracellular membranes and use the RuMP (ribulose 5-phosphate) pathway for carbon assimilation. In contrast, Type II methanotrophs possess intracellular membranes in the cell periphery and use the serine pathway to degrade the formaldehyde produced. Methylococcus capsulatus (Bath) is classified as a Type X methanotroph (1), since it shows the physiological properties of both Type I and II methanotrophs, but it develops Type I intracytoplasmic membranes.Two methane monooxygenases (MMO) 1 catalyze the methane oxidation process in methanotrophs, converting methane to methanol. All methanotrophs express a membrane-bound copper-containing particulate MMO (pMMO) (1, 3), while Type X and a few Type II methanotrophic bacteria are capable of producing a second, soluble form (soluble methane monooxygenase, sMMO). Copper ions are known to switch the methane oxidation from sMMO to pMMO. At low copper-to-biomass ratios, the cytoplasmic sMMO is the dominant MMO. When the cells are grown at high copper-to-biomass ratios, pMMO is expressed and produced instead in the plasma membrane.Because of the unique role of copper ions in regulating the switch between sMMO and pMMO, we have applied cICAT (cleavable isot...