Generic Annotation Research Articles

\u201cGuilt by association\u201d is not competitive with genetic association for identifying autism risk genes

Discovering genes involved in complex human genetic disorders is a major challenge. Many have suggested that machine learning (ML) algorithms using gene networks can be used to supplement traditional genetic association-based approaches to predict or prioritize disease genes. However, questions have been raised about the utility of ML methods for this type of task due to biases within the data, and poor real-world performance. Using autism spectrum disorder (ASD) as a test case, we sought to investigate the question: can machine learning aid in the discovery of disease genes? We collected 13 published ASD gene prioritization studies and evaluated their performance using known and novel high-confidence ASD genes. We also investigated their biases towards generic gene annotations, like number of association publications. We found that ML methods which do not incorporate genetics information have limited utility for prioritization of ASD risk genes. These studies perform at a comparable level to generic measures of likelihood for the involvement of genes in any condition, and do not out-perform genetic association studies. Future efforts to discover disease genes should be focused on developing and validating statistical models for genetic association, specifically for association between rare variants and disease, rather than developing complex machine learning methods using complex heterogeneous biological data with unknown reliability.

An empirical evaluation of electronic annotation tools for Twitter data.

Despite a growing number of natural language processing shared-tasks dedicated to the use of Twitter data, there is currently no ad-hoc annotation tool for the purpose. During the 6th edition of Biomedical Linked Annotation Hackathon (BLAH), after a short review of 19 generic annotation tools, we adapted GATE and TextAE for annotating Twitter timelines. Although none of the tools reviewed allow the annotation of all information inherent of Twitter timelines, a few may be suitable provided the willingness by annotators to compromise on some functionality.

A Generic Human-Machine Annotation Framework Based on Dynamic Cooperative Learning.

The task of obtaining meaningful annotations is a tedious work, incurring considerable costs and time consumption. Dynamic active learning and cooperative learning are recently proposed approaches to reduce human effort of annotating data with subjective phenomena. In this paper, we introduce a novel generic annotation framework, with the aim to achieve the optimal tradeoff between label reliability and cost reduction by making efficient use of human and machine work force. To this end, we use dropout to assess model uncertainty and thereby to decide which instances can be automatically labeled by the machine and which ones require human inspection. In addition, we propose an early stopping criterion based on inter-rater agreement in order to focus human resources on those ambiguous instances that are difficult to label. In contrast to the existing algorithms, the new confidence measures are not only applicable to binary classification tasks but also regression problems. The proposed method is evaluated on the benchmark datasets for non-native English prosody estimation, provided in the Interspeech computational paralinguistics challenge. In the result, the novel dynamic cooperative learning algorithm yields 0.424 Spearman's correlation coefficient compared to 0.413 with passive learning, while reducing the amount of human annotations by 74%.

CardioClassifier: disease- and gene-specific computational decision support for clinical genome interpretation

PurposeInternationally adopted variant interpretation guidelines from the American College of Medical Genetics and Genomics (ACMG) are generic and require disease-specific refinement. Here we developed CardioClassifier (http://www.cardioclassifier.org), a semiautomated decision-support tool for inherited cardiac conditions (ICCs). MethodsCardioClassifier integrates data retrieved from multiple sources with user-input case-specific information, through an interactive interface, to support variant interpretation. Combining disease- and gene-specific knowledge with variant observations in large cohorts of cases and controls, we refined 14 computational ACMG criteria and created three ICC-specific rules. ResultsWe benchmarked CardioClassifier on 57 expertly curated variants and show full retrieval of all computational data, concordantly activating 87.3% of rules. A generic annotation tool identified fewer than half as many clinically actionable variants (64/219 vs. 156/219, Fisher’s P = 1.1 × 10−18), with important false positives, illustrating the critical importance of disease and gene-specific annotations. CardioClassifier identified putatively disease-causing variants in 33.7% of 327 cardiomyopathy cases, comparable with leading ICC laboratories. Through addition of manually curated data, variants found in over 40% of cardiomyopathy cases are fully annotated, without requiring additional user-input data. ConclusionCardioClassifier is an ICC-specific decision-support tool that integrates expertly curated computational annotations with case-specific data to generate fast, reproducible, and interactive variant pathogenicity reports, according to best practice guidelines.

MSeqDR mvTool: A mitochondrial DNA Web and API resource for comprehensive variant annotation, universal nomenclature collation, and reference genome conversion.

Accurate mitochondrial DNA (mtDNA) variant annotation is essential for the clinical diagnosis of diverse human diseases. Substantial challenges to this process include the inconsistency in mtDNA nomenclatures, the existence of multiple reference genomes, and a lack of reference population frequency data. Clinicians need a simple bioinformatics tool that is user-friendly, and bioinformaticians need a powerful informatics resource for programmatic usage. Here, we report the development and functionality of the MSeqDR mtDNA Variant Tool set (mvTool), a one-stop mtDNA variant annotation and analysis Web service. mvTool is built upon the MSeqDR infrastructure (https://mseqdr.org), with contributions of expert curated data from MITOMAP (https://www.mitomap.org) and HmtDB (https://www.hmtdb.uniba.it/hmdb). mvTool supports all mtDNA nomenclatures, converts variants to standard rCRS- and HGVS-based nomenclatures, and annotates novel mtDNA variants. Besides generic annotations from dbNSFP and Variant Effect Predictor (VEP), mvTool provides allele frequencies in more than 47,000 germline mitogenomes, and disease and pathogenicity classifications from MSeqDR, Mitomap, HmtDB and ClinVar (Landrum etal., 2013). mvTools also provides mtDNA somatic variants annotations. "mvTool API" is implemented for programmatic access using inputs in VCF, HGVS, or classical mtDNA variant nomenclatures. The results are reported as hyperlinked html tables, JSON, Excel, and VCF formats. MSeqDR mvTool is freely accessible at https://mseqdr.org/mvtool.php.

The FAUST framework: Free-form annotations on unfolding vascular structures for treatment planning

For complex interventions, such as stenting of a cerebral aneurysm, treatment planning is mandatory. Sketching can support the physician as it involves an active involvement with complex spatial relations and bears a great potential to improve communication. These sketches are employed as direct annotation on 2D medical image data and print outs, respectively. Annotating 3D planning models is more difficult due to possible occlusions of the complex spatial anatomy of vascular structures. Furthermore, the annotations should adapt accordingly to view changes and deforming structures.Therefore, we developed the FAUST framework, which allows creating 3D annotations by freely sketching in the 3D environment. Additionally to generic annotations, the physician is supported to create the most common treatment options with sketching single strokes only. We allow an interactive unfolding of vascular structures with adapting annotations to still convey their meta information. Our framework is realized on the zSpace, which combines a semi-immersive stereoscopic display and a stylus with ray-based interaction techniques.We conducted a user study with computer scientists, carried out a demo session with a neuroradiologist and assessed the performance. The user study revealed a positive rating of the interaction techniques and a high sense of presence. The neuroradiologist stated that our framework can support treatment planning and leads to a better understanding of anatomical structures. Our performance evaluation showed that our sketching approach is usable in real-time with a large number of annotations. Furthermore, our approach can be adapted to a wider range of applications including medical documentation.

Xander: employing a novel method for efficient gene-targeted metagenomic assembly.

BackgroundMetagenomics can provide important insight into microbial communities. However, assembling metagenomic datasets has proven to be computationally challenging. Current methods often assemble only fragmented partial genes.ResultsWe present a novel method for targeting assembly of specific protein-coding genes. This method combines a de Bruijn graph, as used in standard assembly approaches, and a protein profile hidden Markov model (HMM) for the gene of interest, as used in standard annotation approaches. These are used to create a novel combined weighted assembly graph. Xander performs both assembly and annotation concomitantly using information incorporated in this graph. We demonstrate the utility of this approach by assembling contigs for one phylogenetic marker gene and for two functional marker genes, first on Human Microbiome Project (HMP)-defined community Illumina data and then on 21 rhizosphere soil metagenomic datasets from three different crops totaling over 800 Gbp of unassembled data. We compared our method to a recently published bulk metagenome assembly method and a recently published gene-targeted assembler and found our method produced more, longer, and higher quality gene sequences.ConclusionXander combines gene assignment with the rapid assembly of full-length or near full-length functional genes from metagenomic data without requiring bulk assembly or post-processing to find genes of interest. HMMs used for assembly can be tailored to the targeted genes, allowing flexibility to improve annotation over generic annotation pipelines. This method is implemented as open source software and is available at https://github.com/rdpstaff/Xander_assembler.Electronic supplementary materialThe online version of this article (doi:10.1186/s40168-015-0093-6) contains supplementary material, which is available to authorized users.

Personalized image annotation via class-specific cross-domain learning

In this paper, we propose to learn users’ own profiles for image annotation with the purpose of facilitating image searching towards users’ intentions. Considering that the size of a user’s annotation vocabulary is usually small and different users have different visual understanding towards a specific tag, we perform personalization in a class-specific form to summarize a user’s annotation profiles for each tag in his vocabulary. In particular, we first exploit a generic annotation dataset with a class-specific weighted nearest neighbor (cs-WNN) model by combining the techniques of multiple kernel learning and nearest neighbor modeling. Next, a new personalization method, namely class-specific cross-domain learning (cs-CDL), is proposed to achieve users’ own annotation profiles (i.e. the user-specific parameters of cs-WNN models) by exploiting users’ annotation datasets. Experimental results have been reported over several challenging image databases to validate the effectiveness of the proposed method for both generic and personalized image annotation.

The CoMeRe corpus for French : structuring and annotating heterogeneous CMC genres

The CoMeRe project aims to build a kernel corpus of different Computer-Mediated Com-munication (CMC) genres with interactions in French as the main language, by assembling interactions stemming from networks such as the Internet or telecommunication, as well as mono and multimodal, synchronous and asynchronous communications. Corpora are assem-bled using a standard, thanks to the TEI (Text Encoding Initiative) format. This implies extending, through a European endeavor, the TEI model of text, in order to encompass the richest and the more complex CMC genres. This paper presents the Interaction Space model. We explain how this model has been encoded within the TEI corpus header and body. The model is then instantiated through the first four corpora we have processed: three corpora where interactions occurred in single-modality environments (text chat, or SMS systems) and a fourth corpus where text chat, email and forum modalities were used simultaneously. The CoMeRe project has two main research perspectives: Discourse Analysis, only alluded to in this paper, and the linguistic study of idiolects occurring in different CMC genres. As NLP algorithms are an indispensable prerequisite for such research, we present our motiva-tions for applying an automatic annotation process to the CoMeRe corpora. Our wish to guarantee generic annotations meant we did not consider any processing beyond morphosyn-tactic labelling, but prioritized the automatic annotation of any freely variant elements within the corpora. We then turn to decisions made concerning which annotations to make for which units and describe the processing pipeline for adding these. All CoMeRe corpora are verified, thanks to a staged quality control process, designed to allow corpora to move from one project phase to the next. Public release of the CoMeRe corpora is a short-term goal: corpora will be integrated into the forthcoming French National Reference Corpus, and disseminated through the national linguistic infrastructure ORTOLANG. We, therefore, highlight issues and decisions made concerning the OpenData perspective.

Generic Detection and Annotations of the Statically Linked Code

Detection of the statically linked code is one of the important steps in a process of decompilation. It restricts a code, which can be skipped by the decompiler. Type annotations provide an additional information about the number, types, and suitable names for arguments and return values of recognized functions in recognized statically linked code. This is important for generation of calls for these functions. The detection is based on the generic signatures, which are created from the static libraries. The signatures are composed of the first bytes of library modules, CRC codes, module sizes, public symbols, and optionally tail bits or references. A tree structure of signature improves performance by decreasing a count of compared bytes. Generic approach of detection is achieved by an usage of a common object file format. The process is not restricted on specific architecture or file format. However, there are situations when a conflict in the detection can be resolved only by an analysis in the decompiler. Impact of signature usage is verified by the tests with the decompiler.

Quantifying Intrinsic Parallelism Using Linear Algebra for Algorithm/Architecture Coexploration

Degree of parallelism (DoP) is an essential complexity metric that characterizes the number of independent operation sets (IOSs) that can be concurrently executed within an algorithm. This paper presents a generic framework to identify IOSs and to quantify the DoP based on rank theorem in linear algebra. This framework is applied to extract algorithmic parallelisms at various granularities, namely, multigrain parallelism. Our parallelism is intrinsic and platform independent and can provide insights into architectural information, thus facilitating mapping onto generic platforms and early back annotation for modifying algorithms. It plays a significant role in the concurrent optimization of both algorithms and architectures, referred to as Algorithm/Architecture Coexploration (AAC), by trading off between the DoP and the number of operations (NoO). This paper reports three case studies for AAC. The case study on an IDCT reveals that our framework accurately quantifies the parallelism for mapping the algorithm onto generic platforms, including FPGA and multicore systems. The IDCT parallelized by our technique surpasses a conventional spectral parallelization. By exploiting fine-grain parallelism, this paper presents a better porting of a discrete wavelet transform (DWT) onto single instruction multiple data (SIMD) machines compared with a commercial compiler. A high-quality deinterlacer is implemented on a low-cost multicore platform for real-time high-definition applications by analyzing the multigrain parallelism. These case studies reveal the effectiveness of our parallel analysis framework which is applicable to generic systems. Compared with traditional graph traversal techniques, our linear algebraic approach impressively features low complexity and is practical for complicated algorithms.

External Sandhi and its Relevance to Syntactic Treebanking

Externai sandhi is a linguistic phenomenon which refers to a set of sound changes that occur at word boundaries. These changes are similar to phonological processes such as assimilation and fusion when they apply at the level of prosody, such as in connected speech. External sandhi formation can be orthographically reflected in some languages. External sandhi formation in such languages, causes the occurrence of forms which are morphologically unanalyzable, thus posing a problem for all kind of NLP applications. In this paper, we discuss the implications that this phenomenon has for the syntactic annotation of sentences in Telugu, an Indian language with agglutinative morphology. We describe in detail, how external sandhi formation in Telugu, if not handled prior to dependency annotation, leads either to loss or misrepresentation of syntactic information in the treebank. This phenomenon, we argue, necessitates the introduction of a sandhi splitting stage in the generic annotation pipeline currently being followed for the treebanking of Indian languages. We identify one type of external sandhi widely occurring in the previous version of the Telugu treebank (version 0.2) and manually split all its instances leading to the development of a new version 0.5. We also conduct an experiment with a statistical parser to empirically verify the usefulness of the changes made to the treebank. Comparing the parsing accuracies obtained on versions 0. 2 and 0. 5 of the treebank, we observe that splitting even just one type of external sandhi leads to an increase in the overall parsing accuracies

Building neuroinformatic data collections with respect to re-usabilityand sustainability

Event Abstract Back to Event Building neuroinformatic data collections with respect to re-usabilityand sustainability Simon Rerucha1* and Ondrej Kotek1 1 Masaryk University, Faculty of Informatics, Czechia During the research on the field neuroinformatics a large amounts of experimental data are created. These data often contains valuable information that might be re-used withing other research on similar fields, nonetheless the data are often insufficiently organized and described - that factor makes repeated use difficult. Our effort is aimed at reduction of costs to data exploitation ratio, the principal idea is that by spending of certain effort on data organization beyond the scope of their primary use can significantly enhance the value of the data. However the problem is complex and incorporates several aspect, each of which brings different requirements "e.g.data diversity, generic annotation techniques, data sustainability, methods of publication and interchange and finally the motivation of the data sets creators. This contribution proposes a series of methods and recommendation on building reusable and sustainable data collection in the scope of research within Czech National Node for Neuroinformatics (CNNN), but with respect to standards within international neroinformatic community. It is developed on the basis of particular experience with the research within CNNN and the outcomes of the INCF efforts.

Query capabilities of the Karma provenance framework

AbstractProvenance metadata in e‐Science captures the derivation history of data products generated from scientific workflows. Provenance forms a glue linking workflow execution with associated data products, and finds use in determining the quality of derived data, tracking resource usage, and for verifying and validating scientific experiments. In this article, we discuss the scope of provenance collected in the Karma provenance framework used in the LEAD Cyberinfrastructure project, distinguishing provenance metadata from generic annotations. We further describe our approaches to querying for different forms of provenance in Karma in the context of queries in the first provenance challenge. We use an incremental, building‐block method to construct provenance queries based on the fundamental querying capabilities provided by the Karma service centered on the provenance data model. This has the advantage of keeping the Karma service generic and simple, and yet supports a wide range of queries. Karma successfully answers all but one challenge query. Copyright © 2007 John Wiley & Sons, Ltd.