Bayes model

Figure 7. Multiclass NaTve Bayes modeling within Pipeline Pilot software (www.scitegic.com) based on the WOMBAT chemogenomics dataset Probabilistic target predictions are possible for compounds given only their chemical structure. In the example shown, the WOMBAT targets were predicted for Calphostin C, a known protein kinase C inhibiting natural product Tubulin and beta-hexosaminidase are predicted as additional possible targets.

Cao, J., Panetta, R., Yue, S., Steyaert, A., Young-BelKdo, M. and Ahmad, S. (2003) A naive Bayes model to predict coupling between seven transmembrane domain receptors and G-proteins. Bioinformatics 19, 234-240. 

Using a computer-aided model for the prediction of pseu-doallergic reactions from prospective data collected from 581 patients in a controlled clinical trial with an outdated formulation of polygeline, accurate prediction of 86% of the patients who had a systemic reaction was possible (9). The data were handled by multivariate analysis using the independence Bayes model. The predictive accuracy of other reactions was poor. A history of allergy was recorded in 26% of the patients who had systemic reactions and in 12 and 13% of the patients with no systemic or skin reactions. However, these differences were not statistically significant. 

The identification of chemical features that cause compounds to be either selective or promiscuous is essential for predictive safety pharmacology methods to inform medicinal chemistry decisions. Therefore an analysis using Bayesian models along with ECFP descriptors was used. By using the compound annotation from the previous step, a multicategory Bayes model can be trained. The protocol is as follows ... 

Train Bayes model using ECFP descriptors for compounds. 

Cho, H. J., and Lee, J. K. (2008). Error-pooling empirical Bayes model for enhanced statistical discovery of differential expression in microarray data. IEEE Trans. Syst. Man Cybernet. Part A.- SysL Hum., 38(2) 425. 

A naive Bayes classifier is a simple probabilistic classifier based on the so-called Bayes theorem with strong independence assumptions and is particularly suited when the dimensionality of the inputs is high. The naive Bayes model assumes that, given a class r = j, the features X, are independent. Despite its simplicity, the naive Bayes classifier is known to be a robust method even if the independence assumption does not hold (Michalski and Kaufman, 2001). 

Demichelis, R, Magni, R, Piergiorgi, R, Rubin M. A., and Bellazzi, R. (2006). A hierarchical naive Bayes model for handling sample heterogeneity in classification problems An application to tissue micioarrays. BMC Bioinformatics, 7 514. 

Neale, L.C. (1969). AldenResearch Laboratories. Worcester Polytechnic Journal73(1) 11-14. P Neale, L.C. (1971). Chesapeake Bay Model study for Calvert Cliffs. Journal of the Power Division ASCE 97(P04) 827-839. files.asme.org/ASMEORG/... /History/... /5489.pd... 

The new expert s estimates can now be updated using general Bayes model. The mean of this posterior (jx), as the distribution marker, is compared with the true value (jxlvi), in order to determine if and how much the formulated likelihood function has been able to reduced the error of estimates. 

We develop an empirical Bayes model that uses the pooling of observations across all triggers to enhance the analysis. We consider the problems of pool homogenisation, in which data from several pools are re-scaled to belong to a single, narrower pool. In particular, we consider problems relating to the mis-specification of homogenisation factors and its effect on the analysis. 

In this account, we consider the use of empirical Bayes methods for rare events, as discussed in Quigley et al. (2007). Empirical Bayes methods have been applied in the fields of reliability (Sarhan 2003) and risk analysis (Martz et al. 1999) and are regularly used to analyse accident occurrence patterns in road safety applications (Persaud and Lyon 2007). Empirical Bayes models have been shown to perform well against full Bayes models (Camara andTsokos 1999), and even favourably when data is scarce (Vaurio and Jankala 2006, Srivastava and Kubokawa 2007). 

Wang et al. 2012. Cross-subject workload classification with a hierarchical bayes model. Neuroimage 59 (1) 64-69. 

---