Higher order elicitability and Osband’s principle

A statistical functional, such as the mean or the median, is called elicitable if there is a scoring function or loss function such that the correct forecast of the functional is the unique minimizer of the expected score. Such scoring functions are called strictly consistent for the functional. The elicitability of a functional opens the possibility to compare competing forecasts and to rank them in terms of their realized scores. In this paper, we explore the notion of elicitability for multi-dimensional functionals and give both necessary and sufficient conditions for strictly consistent scoring functions. We cover the case of functionals with elicitable components, but we also show that one-dimensional functionals that are not elicitable can be a component of a higher order elicitable functional. In the case of the variance, this is a known result. However, an important result of this paper is that spectral risk measures with a spectral measure with finite support are jointly elicitable if one adds the “correct” quantiles. A direct consequence of applied interest is that the pair (Value at Risk, Expected Shortfall) is jointly elicitable under mild conditions that are usually fulfilled in risk management applications.

MHC class II binding prediction by molecular docking

Proteins of the Major Histocompatibility Complex (MHC) bind self and nonself peptide antigens or epitopes within the cell and present them at the cell surface for recognition by T cells. All T-cell epitopes are MHC binders but not all MCH binders are T-cell epitopes. The MHC class II proteins are extremely polymorphic. Polymorphic residues cluster in the peptide-binding region and largely determine the MHC's peptide selectivity. The peptide binding site on MHC class II proteins consist of five binding pockets. Using molecular docking, we have modelled the interactions between peptide and MHC class II proteins from locus DRB1. A combinatorial peptide library was generated by mutation of residues at peptide positions which correspond to binding pockets (so called anchor positions). The binding affinities were assessed using different scoring functions. The normalized scoring functions for each amino acid at each anchor position were used to construct quantitative matrices (QM) for MHC class II binding prediction. Models were validated by external test sets comprising 4540 known binders. Eighty percent of the known binders are identified in the best predicted 15% of all overlapping peptides, originating from one protein. © 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Top-K Query Processing in Edge-Labeled Graph Data

Edge-labeled graphs have proliferated rapidly over the last decade due to the increased popularity of social networks and the Semantic Web. In social networks, relationships between people are represented by edges and each edge is labeled with a semantic annotation. Hence, a huge single graph can express many different relationships between entities. The Semantic Web represents each single fragment of knowledge as a triple (subject, predicate, object), which is conceptually identical to an edge from subject to object labeled with predicates. A set of triples constitutes an edge-labeled graph on which knowledge inference is performed. Subgraph matching has been extensively used as a query language for patterns in the context of edge-labeled graphs. For example, in social networks, users can specify a subgraph matching query to find all people that have certain neighborhood relationships. Heavily used fragments of the SPARQL query language for the Semantic Web and graph queries of other graph DBMS can also be viewed as subgraph matching over large graphs. Though subgraph matching has been extensively studied as a query paradigm in the Semantic Web and in social networks, a user can get a large number of answers in response to a query. These answers can be shown to the user in accordance with an importance ranking. In this thesis proposal, we present four different scoring models along with scalable algorithms to find the top-k answers via a suite of intelligent pruning techniques. The suggested models consist of a practically important subset of the SPARQL query language augmented with some additional useful features. The first model called Substitution Importance Query (SIQ) identifies the top-k answers whose scores are calculated from matched vertices' properties in each answer in accordance with a user-specified notion of importance. The second model called Vertex Importance Query (VIQ) identifies important vertices in accordance with a user-defined scoring method that builds on top of various subgraphs articulated by the user. Approximate Importance Query (AIQ), our third model, allows partial and inexact matchings and returns top-k of them with a user-specified approximation terms and scoring functions. In the fourth model called Probabilistic Importance Query (PIQ), a query consists of several sub-blocks: one mandatory block that must be mapped and other blocks that can be opportunistically mapped. The probability is calculated from various aspects of answers such as the number of mapped blocks, vertices' properties in each block and so on and the most top-k probable answers are returned. An important distinguishing feature of our work is that we allow the user a huge amount of freedom in specifying: (i) what pattern and approximation he considers important, (ii) how to score answers - irrespective of whether they are vertices or substitution, and (iii) how to combine and aggregate scores generated by multiple patterns and/or multiple substitutions. Because so much power is given to the user, indexing is more challenging than in situations where additional restrictions are imposed on the queries the user can ask. The proposed algorithms for the first model can also be used for answering SPARQL queries with ORDER BY and LIMIT, and the method for the second model also works for SPARQL queries with GROUP BY, ORDER BY and LIMIT. We test our algorithms on multiple real-world graph databases, showing that our algorithms are far more efficient than popular triple stores.

Improving drug discovery using hybrid softcomputing methods

Virtual screening (VS) methods can considerably aid clinical research, predicting how ligands interact with drug targets. Most VS methods suppose a unique binding site for the target, but it has been demonstrated that diverse ligands interact with unrelated parts of the target and many VS methods do not take into account this relevant fact. This problem is circumvented by a novel VS methodology named BINDSURF that scans the whole protein surface in order to find new hotspots, where ligands might potentially interact with, and which is implemented in last generation massively parallel GPU hardware, allowing fast processing of large ligand databases. BINDSURF can thus be used in drug discovery, drug design, drug repurposing and therefore helps considerably in clinical research. However, the accuracy of most VS methods and concretely BINDSURF is constrained by limitations in the scoring function that describes biomolecular interactions, and even nowadays these uncertainties are not completely understood. In order to improve accuracy of the scoring functions used in BINDSURF we propose a hybrid novel approach where neural networks (NNET) and support vector machines (SVM) methods are trained with databases of known active (drugs) and inactive compounds, being this information exploited afterwards to improve BINDSURF VS predictions.

Improvement of Virtual Screening Predictions using Computational Intelligence Methods

Virtual Screening (VS) methods can considerably aid clinical research, predicting how ligands interact with drug targets. However, the accuracy of most VS methods is constrained by limitations in the scoring function that describes biomolecular interactions, and even nowadays these uncertainties are not completely understood. In order to improve accuracy of scoring functions used in most VS methods we propose a hybrid novel approach where neural networks (NNET) and support vector machines (SVM) methods are trained with databases of known active (drugs) and inactive compounds, this information being exploited afterwards to improve VS predictions.

Descubrimiento de fármacos basado en cribado virtual refinado con enfoques neuronales paralelos

En el campo de la medicina clínica es crucial poder determinar la seguridad y la eficacia de los fármacos actuales y además acelerar el descubrimiento de nuevos compuestos activos. Para ello se llevan a cabo ensayos de laboratorio, que son métodos muy costosos y que requieren mucho tiempo. Sin embargo, la bioinformática puede facilitar enormemente la investigación clínica para los fines mencionados, ya que proporciona la predicción de la toxicidad de los fármacos y su actividad en enfermedades nuevas, así como la evolución de los compuestos activos descubiertos en ensayos clínicos. Esto se puede lograr gracias a la disponibilidad de herramientas de bioinformática y métodos de cribado virtual por ordenador (CV) que permitan probar todas las hipótesis necesarias antes de realizar los ensayos clínicos, tales como el docking estructural, mediante el programa BINDSURF. Sin embargo, la precisión de la mayoría de los métodos de CV se ve muy restringida a causa de las limitaciones presentes en las funciones de afinidad o scoring que describen las interacciones biomoleculares, e incluso hoy en día estas incertidumbres no se conocen completamente. En este trabajo abordamos este problema, proponiendo un nuevo enfoque en el que las redes neuronales se entrenan con información relativa a bases de datos de compuestos conocidos (proteínas diana y fármacos), y se aprovecha después el método para incrementar la precisión de las predicciones de afinidad del método de CV BINDSURF.

PokerMapper: : mapping executive functions, poker playing ability and responsible gambling in online environments

Research objectives Poker and responsible gambling both entail the use of the executive functions (EF), which are higher-level cognitive abilities. The main objective of this work was to assess if online poker players of different ability show different performances in their EF and if so, which functions are the most discriminating ones. The secondary objective was to assess if the EF performance can predict the quality of gambling, according to the Gambling Related Cognition Scale (GRCS), the South Oaks Gambling Screen (SOGS) and the Problem Gambling Severity Index (PGSI). Sample and methods The study design consisted of two stages: 46 Italian active players (41m, 5f; age 32±7,1ys; education 14,8±3ys) fulfilled the PGSI in a secure IT web system and uploaded their own hand history files, which were anonymized and then evaluated by two poker experts. 36 of these players (31m, 5f; age 33±7,3ys; education 15±3ys) accepted to take part in the second stage: the administration of an extensive neuropsychological test battery by a blinded trained professional. To answer the main research question we collected all final and intermediate scores of the EF tests on each player together with the scoring on the playing ability. To answer the secondary research question, we referred to GRCS, PGSI and SOGS scores.  We determined which variables that are good predictors of the playing ability score using statistical techniques able to deal with many regressors and few observations (LASSO, best subset algorithms and CART). In this context information criteria and cross-validation errors play a key role for the selection of the relevant regressors, while significance testing and goodness-of-fit measures can lead to wrong conclusions.   Preliminary findings We found significant predictors of the poker ability score in various tests. In particular, there are good predictors 1) in some Wisconsin Card Sorting Test items that measure flexibility in choosing strategy of problem-solving, strategic planning, modulating impulsive responding, goal setting and self-monitoring, 2) in those Cognitive Estimates Test variables related to deductive reasoning, problem solving, development of an appropriate strategy and self-monitoring, 3) in the Emotional Quotient Inventory Short (EQ-i:S) Stress Management score, composed by the Stress Tolerance and Impulse Control scores, and in the Interpersonal score (Empathy, Social Responsibility, Interpersonal Relationship). As for the quality of gambling, some EQ-i:S scales scores provide the best predictors: General Mood for the PGSI; Intrapersonal (Self-Regard; Emotional Self-Awareness, Assertiveness, Independence, Self-Actualization) and Adaptability  (Reality Testing, Flexibility, Problem Solving) for the SOGS, Adaptability for the GRCS. Implications for the field Through PokerMapper we gathered knowledge and evaluated the feasibility of the construction of short tasks/card games in online poker environments for profiling users’ executive functions. These card games will be part of an IT system able to dynamically profile EF and provide players with a feedback on their expected performance and ability to gamble responsibly in that particular moment. The implementation of such system in existing gambling platforms could lead to an effective proactive tool for supporting responsible gambling. 

Comparison of Linear Functions in Middle Grades Textbooks from Singapore and the United States

Many U.S. students do not perform well on mathematics assessments with respect to algebra topics such as linear functions, a building-block for other functions. Poor achievement of U.S. middle school students in this topic is a problem. U.S. eighth graders have had average mathematics scores on international comparison tests such as Third International Mathematics Science Study, later known as Trends in Mathematics and Science Study, (TIMSS)-1995, -99, -03, while Singapore students have had highest average scores. U.S. eighth grade average mathematics scores improved on TIMMS-2007 and held steady onTIMMS-2011. Results from national assessments, PISA 2009 and 2012 and National Assessment of Educational Progress of 2007, 2009, and 2013, showed a lack of proficiency in algebra. Results of curriculum studies involving nations in TIMSS suggest that elementary textbooks in high-scoring countries were different than elementary textbooks and middle grades texts were different with respect to general features in the U.S. The purpose of this study was to compare treatments of linear functions in Singapore and U.S. middle grades mathematics textbooks. Results revealed features currently in textbooks. Findings should be valuable to constituencies who wish to improve U.S. mathematics achievement. Portions of eight Singapore and nine U.S. middle school student texts pertaining to linear functions were compared with respect to 22 features in three categories: (a) background features, (b) general features of problems, and (c) specific characterizations of problem practices, problem-solving competency types, and transfer of representation. Features were coded using a codebook developed by the researcher. Tallies and percentages were reported. Welch's t-tests and chi-square tests were used, respectively, to determine whether texts differed significantly for the features and if codes were independent of country. U.S. and Singapore textbooks differed in page appearance and number of pages, problems, and images. Texts were similar in problem appearance. Differences in problems related to assessment of conceptual learning. U.S. texts contained more problems requiring (a) use of definitions, (b) single computation, (c) interpreting, and (d) multiple responses. These differences may stem from cultural differences seen in attitudes toward education. Future studies should focus on density of page, spiral approach, and multiple response problems.

The Expanded Human Kallikrein (KLK) gene family : genomic organisation, tissue specific expression and potential functions

The tissue kallikreins are serine proteases encoded by highly conserved multigene families. The rodent kallikrein (KLK) families are particularly large, consisting of 13 26 genes clustered in one chromosomal locus. It has been recently recognised that the human KLK gene family is of a similar size (15 genes) with the identification of another 12 related genes (KLK4-KLK15) within and adjacent to the original human KLK locus (KLK1-3) on chromosome 19q13.4. The structural organisation and size of these new genes is similar to that of other KLK genes except for additional exons encoding 5 or 3 untranslated regions. Moreover, many of these genes have multiple mRNA transcripts, a trait not observed with rodent genes. Unlike all other kallikreins, the KLK4-KLK15 encoded proteases are less related (25–44%) and do not contain a conventional kallikrein loop. Clusters of genes exhibit high prostatic (KLK2-4, KLK15) or pancreatic (KLK6-13) expression, suggesting evolutionary conservation of elements conferring tissue specificity. These genes are also expressed, to varying degrees, in a wider range of tissues suggesting a functional involvement of these newer human kallikrein proteases in a diverse range of physiological processes.