In silico toxicity prediction

The modern science of in silico toxicity prediction has made great strides since its inception in 1962 [13]. Nevertheless, there are still many problems to be overcome [107], and it is to be hoped that future work in this essential field will take into account the following recommendations ... 

It has to be emphasized that comprehensive, high-quality datasets for more complex toxicological endpoints like organ toxicities (e.g., liver, kidney), cardiac safety, and teratogenicity are still not really available. Extraction of all relevant data from different sources and structured storage to enable automated data mining and analysis must be the first step. This would be crucial for any further progress in the field of in silico toxicity predictions. 

Lazar (http //lazar.in silico.de/predict) is a k-nearest-neighbor approach to predict chemical endpoints from a training set based on structural fragments [43]. It derives predictions for query structures from a database with experimentally determined toxicity data [43]. Model provides prediction for four endpoints Acute toxicity to fish (lethality) Fathead Minnow Acute Toxicity (LC50), Carcinogenicity, Mutagenicity, and Repeated dose toxicity. 

Onchidal and fasciculins are interesting natural compounds and it is difficult to predict their toxicity. In the case of onchidal, in silico computational predictive modeling for toxic endpoints of interest may prove useful for risk assessment decision support. Likewise, it is a challenge to predict the mditary potential and human impact of these natural toxins since their affinity for enzyme inhibition depends upon the amount and duration of the human exposure. 

Dearden JC. In silico prediction of drng toxicity. J Comput-Aid Mol Des 2003 17 119-27. 

The need to decrease costs and reduce animal suffering for chemical risk assessment has ever more encouraged the use of methods alternative to the use of animals to predict toxicity. These alternative methods can be generally divided into two subgroups study of toxicity in laboratory tubes on small organisms (in vitro) and computational techniques (in silico). 

The use of computational techniques to predict toxicity, or in silico approaches, aims to decrease costs and reduce animal suffering for chemical risk assessment. 

Use in silico systems to predict ADME and toxic properties... 

Elizabeth Barrett Browning after hearing a discussion of toxicity prediction, How can I kill thee, let me count the ways . A recent article by Stouch et al. [74] presents a thoughtful analysis of the validation effort for four such ADME/Tox models. Oprea et al. [75, 76[ have compared drugs leads with compounds in development and in the marketplace and shown that compounds increase in molecular weight and logP as they progress to the bedside. In silico approaches certainly have their place in the pharmaceutical industry as one more tool to increase the probability of success [77]. 

Dobler, M., Till, M.A., and Vedani, A. From crystal stmctures and their analysis to the in silico prediction of toxic phenomena. Helv. Chim. Acta 2003, 86, 1554-1568. 

Dearden, J.C. (2003) In silico prediction of drug toxicity. Journal of Computer-Aided Molecular Design, 17 (2-4), 119-127. 

Early determination of PK properties (absorption, distribution, metabolism, excretion and toxicity, ADMET) has become a fundamental resource of medicinal chemistry in the LO phase. New technologies have been developed to perform a great number of in vitro and even in silico tests. Currently, the most common early-ADME assays evaluate both physicochemical properties (such as the solubility in an opportune medium, the lipophilicity, and the p K i) and biophysical properties (such as the permeability through cellular monolayers to predict oral absorption and the metabolic stability after treatment with liver or microsomal subcellular fraction that contains oxidative cytochromes). 

II. Product Summaries Simulations Plus develops simulation and predictive modeling software for in silico compound screening and for preclinical and clinical drug development in the area of Absorption, Distribution, Metabolism, Excretion and Toxicity (ADMET). The available applications include GastroPlus, ADMET Predictor, ADMET Modeler, DDDPlus, and MembranePlus. 

The two major independent in silico methods for the prediction of toxicity are quantitative-structure-activity-relationship (QSAR) and expert systems (e.g. DEREK, MultiCASE). QSAR means the quantitative relationship between a chemical structure and its biological/ toxicological activity with the help of chemical descriptors that are generated from the... 

There are a number of commercial expert in silico systems available for the prediction of toxicity. Examples of knowledge-based expert systems are DEREK (Lhasa, Leeds, UK), Hazard Expert (ComGenex, San Francisco, CA, USA), Oncologic (San Francisco, CA, USA) and COMPACT (University of Surrey, Guilford, UK). Examples of statistically-based systems are MultiCASE (MultiCase, Beachwood, OH, USA),... 

DfW predictions include information about the mechanism of activity, for example metabolic steps necessary for toxicity. For a more detailed analysis of the toxicity of a compound and its metabolites DfW can be used in combination with an in silico program for the prediction of metabolites. LHASA Limited offers such a program with the name METEOR. For the combination of these two systems, first the metabolites of a compound are predicted by METEOR. These metabolites are then sent to DfW for the prediction of toxicity of the compound and its predicted metabolites. 

Ames BN, Durston WE, Yamasaki E, Lee FD (1973) Carcinogens are mutagens a simple test system combining liver homogenates for activation and bacteria for detection. Proc Natl Acad Sci USA 70 2281-2285 Cunningham AR, Klopman G, Rosenkranz HS (1998) Identification of structural features and associated mechanisms of action for carcinogens in rats. Mutat Res 405 9-27 Dearden JC (2003) In silico prediction of drug toxicity. J Comput Aided Mol Des 17 119-127... 

For noncancer effects the use of PBTD models has elucidated the fundamental mechanisms of toxicological interactions. Such mechanistic knowledge linked with Monte Carlo simulations has initially been employed in in silico toxicology to develop models that predict the toxicity of mixtures in time. The combination of PBTK/TD models for individual compounds with binary PBTK/TD models can be achieved by incorporating key mechanistic knowledge on metabolism inhibitions and interactions through shared enzyme pathways. Simulations of such models can then be compared to experimental data and allow conclusions to be reached about their pharmacokinetics and the likelihood of effects being dose additive. 

---