Modernizing persistence-bioaccumulation-toxicity (PBT) assessment with high throughput animal-free methods.

The assessment of persistence (P), bioaccumulation (B), and toxicity (T) of a chemical is a crucial first step at ensuring chemical safety and is a cornerstone of the European Union's chemicals regulation REACH (Registration, Evaluation, Authorization, and Restriction of Chemicals). Existing methods for PBT assessment are overly complex and cumbersome, have produced incorrect conclusions, and rely heavily on animal-intensive testing. We explore how new-approach methodologies (NAMs) can overcome the limitations of current PBT assessment. We propose two innovative hazard indicators, termed cumulative toxicity equivalents (CTE) and persistent toxicity equivalents (PTE). Together they are intended to replace existing PBT indicators and can also accommodate the emerging concept of PMT (where M stands for mobility). The proposed "toxicity equivalents" can be measured with high throughput in vitro bioassays. CTE refers to the toxic effects measured directly in any given sample, including single chemicals, substitution products, or mixtures. PTE is the equivalent measure of cumulative toxicity equivalents measured after simulated environmental degradation of the sample. With an appropriate panel of animal-free or alternative in vitro bioassays, CTE and PTE comprise key environmental and human health hazard indicators. CTE and PTE do not require analytical identification of transformation products and mixture components but instead prompt two key questions: is the chemical or mixture toxic, and is this toxicity persistent or can it be attenuated by environmental degradation? Taken together, the proposed hazard indicators CTE and PTE have the potential to integrate P, B/M and T assessment into one high-throughput experimental workflow that sidesteps the need for analytical measurements and will support the Chemicals Strategy for Sustainability of the European Union.

BIGCHEM: Challenges and Opportunities for Big Data Analysis in Chemistry.

The increasing volume of biomedical data in chemistry and life sciences requires the development of new methods and approaches for their handling. Here, we briefly discuss some challenges and opportunities of this fast growing area of research with a focus on those to be addressed within the BIGCHEM project. The article starts with a brief description of some available resources for "Big Data" in chemistry and a discussion of the importance of data quality. We then discuss challenges with visualization of millions of compounds by combining chemical and biological data, the expectations from mining the "Big Data" using advanced machine-learning methods, and their applications in polypharmacology prediction and target de-convolution in phenotypic screening. We show that the efficient exploration of billions of molecules requires the development of smart strategies. We also address the issue of secure information sharing without disclosing chemical structures, which is critical to enable bi-party or multi-party data sharing. Data sharing is important in the context of the recent trend of "open innovation" in pharmaceutical industry, which has led to not only more information sharing among academics and pharma industries but also the so-called "precompetitive" collaboration between pharma companies. At the end we highlight the importance of education in "Big Data" for further progress of this area.

Critical assessment of QSAR models of environmental toxicity against Tetrahymena pyriformis: focusing on applicability domain and overfitting by variable selection.

The estimation of the accuracy of predictions is a critical problem in QSAR modeling. The "distance to model" can be defined as a metric that defines the similarity between the training set molecules and the test set compound for the given property in the context of a specific model. It could be expressed in many different ways, e.g., using Tanimoto coefficient, leverage, correlation in space of models, etc. In this paper we have used mixtures of Gaussian distributions as well as statistical tests to evaluate six types of distances to models with respect to their ability to discriminate compounds with small and large prediction errors. The analysis was performed for twelve QSAR models of aqueous toxicity against T. pyriformis obtained with different machine-learning methods and various types of descriptors. The distances to model based on standard deviation of predicted toxicity calculated from the ensemble of models afforded the best results. This distance also successfully discriminated molecules with low and large prediction errors for a mechanism-based model developed using log P and the Maximum Acceptor Superdelocalizability descriptors. Thus, the distance to model metric could also be used to augment mechanistic QSAR models by estimating their prediction errors. Moreover, the accuracy of prediction is mainly determined by the training set data distribution in the chemistry and activity spaces but not by QSAR approaches used to develop the models. We have shown that incorrect validation of a model may result in the wrong estimation of its performance and suggested how this problem could be circumvented. The toxicity of 3182 and 48774 molecules from the EPA High Production Volume (HPV) Challenge Program and EINECS (European chemical Substances Information System), respectively, was predicted, and the accuracy of prediction was estimated. The developed models are available online at http://www.qspr.org site.

FunCat functional inference with belief propagation and feature integration.

Pairwise comparison of sequence data is intensively used for automated functional protein annotation, while graphical models emerge as promising candidates for an integration of various heterogeneous features. We designed a model, termed hRMN that integrates different genomic features and implemented a variant of belief propagation for functional annotation transfer. hRMN allows the assignment of multiple functional categories while avoiding common problems in annotation transfer from heterogeneous datasets, such as an independency of the investigated datasets. We benchmarked this system with large-scale annotation transfer (based on the MIPS FunCat ontology) to proteins of the prokaryotes Bacillus subtilis, Helicobacter pylori, Listeria monocytogenes, and Listeria innocua. hRMN consistently outperformed two competitors in annotation of four bacterial genomes. The developed code is available for download at http://mips.gsf.de/proj/bfab/hRMN.html.

Surrogate data--a secure way to share corporate data.

The privacy of chemical structure is of paramount importance for the industrial sector, in particular for the pharmaceutical industry. At the same time, companies handle large amounts of physico-chemical and biological data that could be shared in order to improve our molecular understanding of pharmacokinetic and toxicological properties, which could lead to improved predictivity and shorten the development time for drugs, in particular in the early phases of drug discovery. The current study provides some theoretical limits on the information required to produce reverse engineering of molecules from generated descriptors and demonstrates that the information content of molecules can be as low as less than one bit per atom. Thus theoretically just one descriptor can be used to completely disclose the molecular structure. Instead of sharing descriptors, we propose to share surrogate data. The sharing of surrogate data is nothing else but sharing of reliably predicted molecules. The use of surrogate data can provide the same information as the original set. We consider the practical application of this idea to predict lipophilicity of chemical compounds and we demonstrate that surrogate and real (original) data provides similar prediction ability. Thus, our proposed strategy makes it possible not only to share descriptors, but also complete collections of surrogate molecules without the danger of disclosing the underlying molecular structures.

Super paramagnetic clustering of protein sequences.

BACKGROUND: Detection of sequence homologues represents a challenging task that is important for the discovery of protein families and the reliable application of automatic annotation methods. The presence of domains in protein families of diverse function, inhomogeneity and different sizes of protein families create considerable difficulties for the application of published clustering methods. RESULTS: Our work analyses the Super Paramagnetic Clustering (SPC) and its extension, global SPC (gSPC) algorithm. These algorithms cluster input data based on a method that is analogous to the treatment of an inhomogeneous ferromagnet in physics. For the SwissProt and SCOP databases we show that the gSPC improves the specificity and sensitivity of clustering over the original SPC and Markov Cluster algorithm (TRIBE-MCL) up to 30%. The three algorithms provided similar results for the MIPS FunCat 1.3 annotation of four bacterial genomes, Bacillus subtilis, Helicobacter pylori, Listeria innocua and Listeria monocytogenes. However, the gSPC covered about 12% more sequences compared to the other methods. The SPC algorithm was programmed in house using C++ and it is available at http://mips.gsf.de/proj/spc. The FunCat annotation is available at http://mips.gsf.de. CONCLUSION: The gSPC calculated to a higher accuracy or covered a larger number of sequences than the TRIBE-MCL algorithm. Thus it is a useful approach for automatic detection of protein families and unsupervised annotation of full genomes.

Application of ALOGPS 2.1 to predict log D distribution coefficient for Pfizer proprietary compounds.

Evaluation of the ALOGPS, ACD Labs LogD, and PALLAS PrologD suites to calculate the log D distribution coefficient resulted in high root-mean-squared error (RMSE) of 1.0-1.5 log for two in-house Pfizer's log D data sets of 17,861 and 640 compounds. Inaccuracy in log P prediction was the limiting factor for the overall log D estimation by these algorithms. The self-learning feature of the ALOGPS (LIBRARY mode) remarkably improved the accuracy in log D prediction, and an rmse of 0.64-0.65 was calculated for both data sets.

Application of ALOGPS to predict 1-octanol/water distribution coefficients, logP, and logD, of AstraZeneca in-house database.

The ALOGPS 2.1 was developed to predict 1-octanol/water partition coefficients, logP, and aqueous solubility of neutral compounds. An exclusive feature of this program is its ability to incorporate new user-provided data by means of self-learning properties of Associative Neural Networks. Using this feature, it calculated a similar performance, RMSE = 0.7 and mean average error 0.5, for 2569 neutral logP, and 8122 pH-dependent logD(7.4), distribution coefficients from the AstraZeneca "in-house" database. The high performance of the program for the logD(7.4) prediction looks surprising, because this property also depends on ionization constants pKa. Therefore, logD(7.4) is considered to be more difficult to predict than its neutral analog. We explain and illustrate this result and, moreover, discuss a possible application of the approach to calculate other pharmacokinetic and biological activities of chemicals important for drug development.