Search

The rapid emergence of pesticide resistance has given rise to a demand for herbicides with new mode of action (MoA). In the agrochemical sector, with the availability of experimental high throughput screening (HTS) data, it is now possible to utilize in silico target prediction methods in the early discovery phase to suggest the MoA of a compound via data mining of bioactivity data. While having been established in the pharmaceutical context, in the agrochemical area this approach poses rather different challenges, as we have found in this work, partially due to different chemistry, but even more so due to different (usually smaller) amounts of data, and different ways of conducting HTS. With the aim to apply computational methods for facilitating herbicide target identification, 48,000 bioactivity data against 16 herbicide targets were processed to train Laplacian modified Naïve Bayesian (NB) classification models. The herbicide target prediction model ("HerbiMod") is an ensemble of 16 binary classification models which are evaluated by internal, external and prospective validation sets. In addition to the experimental inactives, 10,000 random agrochemical inactives were included in the training process, which showed to improve the overall balanced accuracy of our models up to 40%. For all the models, performance in terms of balanced accuracy of≥80% was achieved in five-fold cross validation. Ranking target predictions was addressed by means of z-scores which improved predictivity over using raw scores alone. An external testset of 247 compounds from ChEMBL and a prospective testset of 394 compounds from BASF SE tested against five well studied herbicide targets (ACC, ALS, HPPD, PDS and PROTOX) were used for further validation. Only 4% of the compounds in the external testset lied in the applicability domain and extrapolation (and correct prediction) was hence impossible, which on one hand was surprising, and on the other hand illustrated the utilization of using applicability domains in the first place. However, performance better than 60% in balanced accuracy was achieved on the prospective testset, where all the compounds fell within the applicability domain, and which hence underlines the possibility of using target prediction also in the area of agrochemicals.

Chemical structure data and corresponding measured bioactivities of compounds are nowadays easily available from public and commercial databases. However, these databases contain heterogeneous data from different laboratories determined under different protocols and, in addition, sometimes even erroneous entries. In this study, we evaluated the use of data from bioactivity databases for the generation of high quality in silico models for off-target mediated toxicity as a decision support in early drug discovery and crop-protection research. We chose human acetylcholinesterase (hAChE) inhibition as an exemplary end point for our case study. A standardized and thorough quality management routine for input data consisting of more than 2,200 chemical entities from bioactivity databases was established. This procedure finally enables the development of predictive QSAR models based on heterogeneous in vitro data from multiple laboratories. An extended applicability domain approach was used, and regression results were refined by an error estimation routine. Subsequent classification augmented by special consideration of borderline candidates leads to high accuracies in external validation achieving correct predictive classification of 96%. The standardized process described herein is implemented as a (semi)automated workflow and thus easily transferable to other off-targets and assay readouts.

Motivation: Biomedical ontologies have proved to be valuable tools for data analysis and data interoperability. Protein–ligand interactions are key players in drug discovery and development; however, existing public ontologies that describe the knowledge space of biomolecular interactions do not cover all aspects relevant to pharmaceutical modelling and simulation. Results: The protein–ligand interaction ontology (PLIO) was developed around three main concepts, namely target, ligand and interaction, and was enriched by adding synonyms, useful annotations and references. The quality of the ontology was assessed based on structural, functional and usability features. Validation of the lexicalized ontology by means of natural language processing (NLP)-based methods showed a satisfactory performance (F-score = 81%). Through integration into our information retrieval environment we can demonstrate that PLIO supports lexical search in PubMed abstracts. The usefulness of PLIO is demonstrated by two use-case scenarios and it is shown that PLIO is able to capture both confirmatory and new knowledge from simulation and empirical studies. Availability: The PLIO ontology is made freely available to the public at http://www.scai.fraunhofer.de/bioinformatics/downloads.html. Contact: martin.hofmann-apitius@scai.fraunhofer.de Supplementary Information: Supplementary data are available at Bioinformatics online.

The first step in finding a "drug" is screening chemical compound databases against a protein target. In silico approaches like virtual screening by molecular docking are well established in modern drug discovery. As molecular databases of compounds and target structures are becoming larger and more and more computational screening approaches are available, there is an increased need in compute power and more complex workflows. In this regard, computational Grids are predestined and offer seamless compute and storage capacity. In recent projects related to pharmaceutical research, the high computational and data storage demands of large-scale in silico drug discovery approaches have been addressed by using Grid computing infrastructures, in both; pharmaceutical industry as well as academic research. Grid infrastructures are part of the so-called eScience paradigm, where a digital infrastructure supports collaborative processes by providing relevant resources and tools for data- and compute-intensive applications. Substantial computing resources, large data collections and services for data analysis are shared on the Grid infrastructure and can be mobilized on demand. This review gives an overview on the use of Grid computing for in silico drug discovery and tries to provide a vision of future development of more complex and integrated workflows on Grids, spanning from target identification and target validation via protein-structure and ligand dependent screenings to advanced mining of large scale in silico experiments.

Since the development of the first docking algorithm in the early 1980s a variety of different docking approaches and tools has been created in order to solve the docking problem. Subsequent studies have shown that the docking performance of most tools strongly depends on the considered target. Thus it is hard to choose the best algorithm in the situation at hand. The docking tools FlexX and AutoDock are among the most popular programs for docking flexible ligands into target proteins. Their analysis, comparison, and combination are the topics of this study. In contrast to standard consensus scoring techniques which integrate different scoring algorithms usually only by their rank, we focus on a more general approach. Our new combined docking workflow-AutoxX-unifies the interaction models of AutoDock and FlexX rather than combining the scores afterward which allows interpretability of the results. The performance of FlexX, AutoDock, and the combined algorithm AutoxX was evaluated on the basis of a test set of 204 structures from the Protein Data Bank (PDB). AutoDock and FlexX show a highly diverse redocking accuracy at the different complexes which assures again the usefulness of taking several docking algorithms into account. With the combined docking the number of complexes reproduced below an rmsd of 2.5 angstrom could be raised by 10. AutoxX had a strong positive effect on several targets. The highest performance increase could be found when redocking 20 protein-ligand complexes of alpha-thrombin, plasmepsin, neuraminidase, and D-xylose isomerase. A decrease was found for gamma-chymotrypsin. The results show that-applied to the right target-AutoxX can improve the docking performance compared to AutoDock and FlexX alone.

Enabling the seamless integration between applications executing on heterogeneous Grid middleware poses a number of challenges to both application scientists and middleware developers. We highlight some of these challenges in the context of designing and implementing DockFlow. DockFlow is a virtual screening environment integrating four Grid-based protein docking tools that execute on different Grid middleware technologies at different locations. We propose extensions that can be applied to any Grid-based workflow system to support the run-time interoperability between the available tools. The extensions are generic, and as an example we describe how they have been implemented in the InforSense workflow system. We also present experimental results that evaluate the tradeoffs between performance and usability of the proposed methods.