Search

Abstract Recently, imputation techniques have been adapted to predict activity values among sparse bioactivity matrices, showing improvements in predictive performance over traditional QSAR models. These models are able to use experimental activity values for auxiliary assays when predicting the activity of a test compound on a specific assay. In this study, we tested three different multi-task imputation techniques on three classification-based toxicity datasets: two of small scale (12 assays each) and one large scale with 417 assays. Moreover, we analyzed in detail the improvements shown by the imputation models. We found that test compounds that were dissimilar to training compounds, as well as test compounds with a large number of experimental values for other assays, showed the largest improvements. We also investigated the impact of sparsity on the improvements seen as well as the relatedness of the assays being considered. Our results show that even a small amount of additional information can provide imputation methods with a strong boost in predictive performance over traditional single task and multi-task predictive models.

Abstract There has been a growing interest in multitask prediction in chemoinformatics, helped by the increasing use of deep neural networks in this field. This technique is applied to multitarget data sets, where compounds have been tested against different targets, with the aim of developing models to predict a profile of biological activities for a given compound. However, multitarget data sets tend to be sparse; i.e., not all compound-target combinations have experimental values. There has been little research on the effect of missing data on the performance of multitask methods. We have used two complete data sets to simulate sparseness by removing data from the training set. Different models to remove the data were compared. These sparse sets were used to train two different multitask methods, deep neural networks and Macau, which is a Bayesian probabilistic matrix factorization technique. Results from both methods were remarkably similar and showed that the performance decrease because of missing data is at first small before accelerating after large amounts of data are removed. This work provides a first approximation to assess how much data is required to produce good performance in multitask prediction exercises.

Introduction. This paper reports preliminary research in a primarily experimental study of how the general public search for information on the Web. The focus is on the query transformation patterns that characterise searching. Method. In this work, we have used transaction logs from the Excite search engine to develop methods for analysing query transformations that should aid the analysis of our ongoing experimental work. Our methods involve the use of similarity techniques to link queries with the most similar previous query in a train. The resulting query transformations are represented as a list of codes representing a whole search. Analysis. It is shown how query transformation sequences can be represented as graphical networks and some basic statistical results are shown. A correlation analysis is performed to examine the co-occurrence of Boolean and quotation mark changes with the syntactic changes. Results. A frequency analysis of the occurrence of query transformation codes is presented. The connectivity of graphs obtained from the query transformation is investigated and found to follow an exponential scaling law. The correlation analysis reveals a number of patterns that provide some interesting insights into Web searching by the general public. Conclusion. We have developed analytical methods based on query similarity that can be applied to our current experimental work with volunteer subjects. The results of these will form part of a database with the aim of developing an improved understanding of how the public search the Web.

Adgrg6 (Gpr126) is an adhesion class G protein-coupled receptor with a conserved role in myelination of the peripheral nervous system. In the zebrafish, mutation of adgrg6 also results in defects in the inner ear: otic tissue fails to down-regulate versican gene expression and morphogenesis is disrupted. We have designed a whole-animal screen that tests for rescue of both up- and down-regulated gene expression in mutant embryos, together with analysis of weak and strong alleles. From a screen of 3120 structurally diverse compounds, we have identified 68 that reduce versican b expression in the adgrg6 mutant ear, 41 of which also restore myelin basic protein gene expression in Schwann cells of mutant embryos. Nineteen compounds unable to rescue a strong adgrg6 allele provide candidates for molecules that may interact directly with the Adgrg6 receptor. Our pipeline provides a powerful approach for identifying compounds that modulate GPCR activity, with potential impact for future drug design.

Evolutionary algorithms have been widely adopted for solving many of the complex tasks involved in drug design. Here, applications to two different tasks are reviewed: combinatorial library design and deriving quantitative structure-activity relationship (QSAR) models. Combinatorial libraries are the result of combinatorial synthesis whereby robotics is used to synthesise large numbers of compounds in parallel. Many more compounds could be made than can be handled in practice; thus, methods are required for selecting appropriate subsets of compounds. The large numbers involved make this problem well suited to the application of optimisation techniques such as evolutionary algorithms. QSARs attempt to relate a numerical description of molecular structure to known biological activity. Large numbers of easily computable molecular descriptors are now available that could be used to derive models; however, when more descriptors are available than observations, or compounds, overfitting of the data can result where the models generated have poor predictive ability. Thus evolutionary algorithms have successfully been applied to select descriptors that give rise to good QSAR models. [ABSTRACT FROM AUTHOR]