Kurt De Grave

kLog: a language for logical and relational learning with kernels

Abstract We introduce kLog, a novel approach to statistical relational learning. Unlike standard approaches, kLog does not represent a probability distribution directly. It is rather a language to perform kernel-based learning on expressive logical and relational representations. kLog allows users to specify learning problems declaratively. It builds on simple but powerful concepts: learning from interpretations, entity/relationship data modeling, logic programming, and deductive databases. Access by the kernel to the rich representation is mediated by a technique we call graphicalization : the relational representation is first transformed into a graph — in particular, a grounded entity/relationship diagram. Subsequently, a choice of graph kernel defines the feature space. kLog supports mixed numerical and symbolic data, as well as background knowledge in the form of Prolog or Datalog programs as in inductive logic programming systems. The kLog framework can be applied to tackle the same range of tasks that has made statistical relational learning so popular, including classification, regression, multitask learning, and collective classification. We also report about empirical comparisons, showing that kLog can be either more accurate, or much faster at the same level of accuracy, than Tilde and Alchemy. kLog is GPLv3 licensed and is available at http://klog.dinfo.unifi.it along with tutorials.

Machine learning applications in proteomics research: How the past can boost the future

Machine learning is a subdiscipline within artificial intelligence that focuses on algorithms that allow computers to learn solving a (complex) problem from existing data. This ability can be used to generate a solution to a particularly intractable problem, given that enough data are available to train and subsequently evaluate an algorithm on. Since MS‐based proteomics has no shortage of complex problems, and since publicly available data are becoming available in ever growing amounts, machine learning is fast becoming a very popular tool in the field. We here therefore present an overview of the different applications of machine learning in proteomics that together cover nearly the entire wet‐ and dry‐lab workflow, and that address key bottlenecks in experiment planning and design, as well as in data processing and analysis.

Predicting human olfactory perception from chemical features of odor molecules

How will this molecule smell? We still do not understand what a given substance will smell like. Keller et al. launched an international crowd-sourced competition in which many teams tried to solve how the smell of a molecule will be perceived by humans. The teams were given access to a database of responses from subjects who had sniffed a large number of molecules and been asked to rate each smell across a range of different qualities. The teams were also given a comprehensive list of the physical and chemical features of the molecules smelled. The teams produced algorithms to predict the correspondence between the quality of each smell and a given molecule. The best models that emerged from this challenge could accurately predict how a new molecule would smell. Science, this issue p. 820 Results of a crowdsourcing competition show that it is possible to accurately predict and reverse-engineer the smell of a molecule. It is still not possible to predict whether a given molecule will have a perceived odor or what olfactory percept it will produce. We therefore organized the crowd-sourced DREAM Olfaction Prediction Challenge. Using a large olfactory psychophysical data set, teams developed machine-learning algorithms to predict sensory attributes of molecules based on their chemoinformatic features. The resulting models accurately predicted odor intensity and pleasantness and also successfully predicted 8 among 19 rated semantic descriptors (“garlic,” “fish,” “sweet,” “fruit,” “burnt,” “spices,” “flower,” and “sour”). Regularized linear models performed nearly as well as random forest–based ones, with a predictive accuracy that closely approaches a key theoretical limit. These models help to predict the perceptual qualities of virtually any molecule with high accuracy and also reverse-engineer the smell of a molecule.

Cheaper faster drug development validated by the repositioning of drugs against neglected tropical diseases

There is an urgent need to make drug discovery cheaper and faster. This will enable the development of treatments for diseases currently neglected for economic reasons, such as tropical and orphan diseases, and generally increase the supply of new drugs. Here, we report the Robot Scientist ‘Eve’ designed to make drug discovery more economical. A Robot Scientist is a laboratory automation system that uses artificial intelligence (AI) techniques to discover scientific knowledge through cycles of experimentation. Eve integrates and automates library-screening, hit-confirmation, and lead generation through cycles of quantitative structure activity relationship learning and testing. Using econometric modelling we demonstrate that the use of AI to select compounds economically outperforms standard drug screening. For further efficiency Eve uses a standardized form of assay to compute Boolean functions of compound properties. These assays can be quickly and cheaply engineered using synthetic biology, enabling more targets to be assayed for a given budget. Eve has repositioned several drugs against specific targets in parasites that cause tropical diseases. One validated discovery is that the anti-cancer compound TNP-470 is a potent inhibitor of dihydrofolate reductase from the malaria-causing parasite Plasmodium vivax.

Active Learning for High Throughput Screening

An important task in many scientific and engineering disciplines is to set up experiments with the goal of finding the best instances (substances, compositions, designs) as evaluated on an unknown target function using limited resources. We study this problem using machine learning principles, and introduce the novel task of active k-optimization. The problem consists of approximating the kbest instances with regard to an unknown function and the learner is active, that is, it can present a limited number of instances to an oracle for obtaining the target value. We also develop an algorithm based on Gaussian processes for tackling active k-optimization, and evaluate it on a challenging set of tasks related to structure-activity relationship prediction.

Representation of probabilistic scientific knowledge.

The theory of probability is widely used in biomedical research for data analysis and modelling. In previous work the probabilities of the research hypotheses have been recorded as experimental metadata. The ontology HELO is designed to support probabilistic reasoning, and provides semantic descriptors for reporting on research that involves operations with probabilities. HELO explicitly links research statements such as hypotheses, models, laws, conclusions, etc. to the associated probabilities of these statements being true. HELO enables the explicit semantic representation and accurate recording of probabilities in hypotheses, as well as the inference methods used to generate and update those hypotheses. We demonstrate the utility of HELO on three worked examples: changes in the probability of the hypothesis that sirtuins regulate human life span; changes in the probability of hypotheses about gene functions in the S. cerevisiae aromatic amino acid pathway; and the use of active learning in drug design (quantitative structure activity relation learning), where a strategy for the selection of compounds with the highest probability of improving on the best known compound was used. HELO is open source and available at https://github.com/larisa-soldatova/HELO

Molecular Graph Augmentation with Rings and Functional Groups

Molecular graphs are a compact representation of molecules but may be too concise to obtain optimal generalization performance from graph-based machine learning algorithms. Over centuries, chemists have learned what are the important functional groups in molecules. This knowledge is normally not manifest in molecular graphs. In this paper, we introduce a simple method to incorporate this type of background knowledge: we insert additional vertices with corresponding edges for each functional group and ring structure identified in the molecule. We present experimental evidence that, on a wide range of ligand-based tasks and data sets, the proposed augmentation method improves the predictive performance over several graph kernel-based quantitative structure-activity relationship models. When the augmentation technique is used with the recent pairwise maximal common subgraphs kernel, we achieve a significant improvement over the current state-of-the-art on the NCI-60 cancer data set in 28 out of 60 cell lines, with the other 32 cell lines showing no significant difference in accuracy. Finally, on the Bursi mutagenicity data set, we obtain near-optimal predictions.

Designing biomedical proteomics experiments: state-of-the-art and future perspectives

ABSTRACT With the current expanded technical capabilities to perform mass spectrometry-based biomedical proteomics experiments, an improved focus on the design of experiments is crucial. As it is clear that ignoring the importance of a good design leads to an unprecedented rate of false discoveries which would poison our results, more and more tools are developed to help researchers designing proteomic experiments. In this review, we apply statistical thinking to go through the entire proteomics workflow for biomarker discovery and validation and relate the considerations that should be made at the level of hypothesis building, technology selection, experimental design and the optimization of the experimental parameters.

Multi-objective optimization with surrogate trees

Multi-objective optimization problems are usually solved with evolutionary algorithms when the objective functions are cheap to compute, or with surrogate-based optimizers otherwise. In the latter case, the objective functions are modeled with powerful non-linear model learners such as Gaussian Processes or Support Vector Machines, for which the training time can be prohibitively large when dealing with optimization problems with moderately expensive objective functions. In this paper, we investigate the use of model trees as an alternative kind of model, providing a good compromise between high expressiveness and low training time. We propose a fast surrogate-based optimizer exploiting the structure of model trees for candidate selection. The empirical results show the promise of the approach for problems on which classical surrogate-based optimizers are painfully slow.

kLogNLP: Graph Kernel--based Relational Learning of Natural Language

kLog is a framework for kernel-based learning that has already proven successful in solving a number of relational tasks in natural language processing. In this paper, we present kLogNLP, a natural language processing module for kLog. This module enriches kLog with NLP-specific preprocessors, enabling the use of existing libraries and toolkits within an elegant and powerful declarative machine learning framework. The resulting relational model of the domain can be extended by specifying additional relational features in a declarative way using a logic programming language. This declarative approach offers a flexible way of experimentation and a way to insert domain knowledge.