Probabilistic screening

Hamed M. M., Conte J. P., and Bedient P. B. (1995) Probabilistic screening tool for ground-water contamination assessment. J. Environ. Eng. 121, 767-775. 

Duvall SE, Barron MG. 2000. A screening level probabilistic risk assessment of mercury in Florida Everglades food webs. Ecotoxicol Environ Saf 47 298-305. 

It is imperative, however, to understand the probabilistic nature of such experiments a promising profile on pharmacogenomic and toxicogenomic screens will enhance the likelihood of having selected an ultimately successful compound, and will achieve this goal quicker than conventional animal experimentation, but will do so only with a certain likelihood of success. The less reductionist approach of the animal experiment will still be needed. It is to be anticipated, however, that such approaches will constitute an important, time- and resource-saving first evaluation or screening step that will help to focus and reduce the number of animal experiments that will ultimately need to be conducted. 

Sanderson H, Johnson DJ, Wilson CJ, Brain RA, Solomon KR (2003) Probabilistic hazard assessment of environmentally occurring pharmaceuticals toxicity to fish, Daphnia and algae by ECOSAR screening. Toxicol Lett 144 383-395... 

Circular substructures of various sorts have been widely used for applications such as structure and substructure searching, constitutional symmetry, structure elucidation and the probabilistic modeling of bioactivity inter alia The work reported here demonstrates that this type of fragment is also very well suited to virtual screening using multiple reference structures. 

Hernandez-Caraballo et al. [91,92] evaluated several classical chemometric methods and ANNs as screening tools for cancer research. They measured the concentrations of Zn, Cu, Fe and Se in blood serum specimens by total reflection XRF spectrometry. The classical chemometric approaches used were PCA and logistic regression. On the other hand, two neural networks were employed for the same task, viz., back-propagation and probabilistic neural networks. 

Current generally applicable biodegradation models focus on the estimation of readily and nonreadily biodegradability in screening tests. This is because most experimental data are from such tests (e.g., MITI-I). There are far fewer data that are both quantitative and environmentally relevant (i.e., measured half-lives or rate constants). However, individual transformations and pathways are well documented in the literature. This allows for development of explicitly mechanistic models, making use of established group-contribution approaches, hierarchic rule-based expert systems, and probabilistic evaluation of possible transformation pathways. 

Although probabilistic predictions are acceptable in virtual screening experiments for lead identification, in lead optimization studies, where only a limited number of compounds is examined, confidence in docked binding modes must be very high. Therefore, the conclusion from the neuraminidase... 

A theoretical approach to determine the required number of screened mutants is based on the landscape paradigm. The number of uphill paths on the landscape given the fitness of the sequence is related to the library size required to probabilistically capture a single path. Using the uncoupled fitness function, Aita and Husimi (1996) analytically... 

The level of detail of the assessment may vary greatly, depending on the purpose for which it is carried out (i.e. as a screening-level assessment or as a detailed, probabilistic assessment). The level of detail with which the uncertainty is analysed will vary accordingly and should, as a rule, be consistent with the level of detail of the exposure assessment. 

An uncertainty analysis gives the assessor the opportunity to re-evaluate the scenario, model approaches and parameters of the analysis and to consider their influence in the overall analysis. The practical impact of uncertainty analysis is illustrated within the annexed case-studies, which also clarify how uncertainty analyses follow a systematic methodology, based on a tiered approach, and consider all possible sources of uncertainty. The first step in uncertainty analysis consists of a screening, followed by a qualitative analysis and two levels of quantitative analysis, using deterministic and probabilistic data. The assessor should be aware that an uncertainty analysis cannot answer all the questions, which, moreover, may lead to new questions. 

Andres, T. H. and Hajas, W.C. (1993). Using iterated fractional factorial design to screen parameters in sensitivity analysis of a probabilistic risk assessment model. Proceedings of the Joint International Conference on Mathematical Models and Supercomputing in... 

One more way to conduct a similarity-based virtual screening is to retrieve the structures containing a user-defined set of pharmacophoric features. In the Dynamic Mapping of Consensus positions (DMC) algorithm those features are selected by finding common positions in bit strings for all active compounds. The potency-scaled DMC algorithm (POT-DMC) " is a modification of DMC in which compounds activities are taken into account. The latter two methods may be considered as intermediate between conventional similarity search and probabilistic SAR approaches. 

Simplistic and heuristic similarity-based approaches can hardly produce as good predictive models as modern statistical and machine learning methods that are able to assess quantitatively biological or physicochemical properties. QSAR-based virtual screening consists of direct assessment of activity values (numerical or binary) of all compounds in the database followed by selection of hits possessing desirable activity. Mathematical methods used for models preparation can be subdivided into classification and regression approaches. The former decide whether a given compound is active, whereas the latter numerically evaluate the activity values. Classification approaches that assess probability of decisions are called probabilistic. 

Numerous studies have been devoted to classification (probabilistic) approaches used in conjunction with fragment descriptors for virtual screening. Here we present several examples. 

Harper et al. have demonstrated a much better performance of probabilistic binary kernel discrimination method to screen large databases compared to... 

In this chapter we overview some probabilistic methods used for biological activity prediction, paying particular attention to the problems of creation of the training and evaluation sets, validation of (Q)SAR models, estimation of prediction accuracy, interpretation of the prediction results and their application in virtual screening. 

Probabilistic Ligand-based Virtual Screening Methods... 

Many different methods can be applied to virtual screening, and such methods are described in other chapters of this book and/or in the Handbooks of Che-minformatics Here we discuss the methods based on a probabilistic approach. Unfortunately, there are many publications in which the probabilistic or statistical approach items are farfetched. The Binary Kernel Discrimination and the Bayesian Machine Learning Models are actually special... 

Niculescu SP, Atkinson A, Hammond G, Lewis M. Using fragment chemistry data mining and probabilistic neural networks in screening chemicals for acute toxicity to the fathead minnow. SAR QSAR Environ Res 2004 15(4) 293-309. 

---