Random Forest

Svetnik V, Liaw A, Tong C, Culberson JC, Sheridan RP, Feuston BP. Random forest a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci 2003 43(6) 1947-58. 

Woolfitt, A. Moura, H. Barr, J. De, B. Popovic,T. Satten, G. Jarman, K. H. Wahl, K. L., Differentiation of Bacillus spp. by MALDI-TOF mass spectrometry using a bacterial fingerprinting algorithm and a random forest classification algorithm. Presented at 5th ISIAM Meeting, Richland, WA 2004. 

Of the physicochemical descriptors, lipophilicity (as described by clogP and Topological Polar Surface Area (TPSA) gave the strongest overall correlation to incidence of adverse in vivo outcomes, whether analyzed in terms of free or total drug threshold concentrations. In the case of free drug threshold analysis, a Random Forest statistical method indicated that there was a higher chance of a compound with TPSA <70... 

Random Forest model with key properties) (Random Forest model with key properties)... 

Figure 1 Random Forest models of dogP and TPSA against relative risk.

Random forest VD is a function of specific chemical attributes Computed chemical descriptors [40]... 

Hao M. Li Y. Wang Y. Zhang S. A classification study of respiratory syncytial virus (RSV) inhibitors by variable selection with random forest. International Journal of Molecular Sciences, 2011, 12 (2), 1259-1280. 

The field emission properties of carbon nanotube forests and single nanotubes are described. Controlled emission is possible for aligned CNT arrays where the spacing is twice the CNT height, as grown by plasma enhanced chemical vapor deposition. This leads to the maximum field enhancement factor. For random forests, the field enhancement obeys an exponential distribution, leading to a lower emission site density and imperfect current sharing. Ballast resistors can help alleviate this problem. Random nanocarbons perform less well than CNTs. Some applications are covered. Elec-... 

Miller, M.D. and Zhang, Y. (2006) A hybrid mixture discriminant analysis-random forest computational model for the prediction of volume of distribution in human. Journal of Medicinal Chemistry,... 

In this study, a machine learning model system was developed to classify cell line chemosensitivity exclusively based on proteomic profiling. Using reverse-phase protein lysate microarrays, protein expression levels were measured by 52 antibodies in a panel of 60 human cancer cell (NCI-60) lines. The model system combined several well-known algorithms, including Random forests, Relief, and the nearest neighbor methods, to construct the protein expression-based chemosensitivity classifiers. 

It is interesting to note that various QSAR/QSPR models from an array of methods can be very different in both complexity and predictivity. For example, a simple QSPR equation with three parameters can predict logP within one unit of measured values (43) while a complex hybrid mixture discriminant analysis-random forest model with 31 computed descriptors can only predict the volume of distribution of drugs in humans within about twofolds of experimental values (44). The volume of distribution is a more complex property than partition coefficient. The former is a physiological property and has a much higher uncertainty in its experimental measurements while logP is a much simpler physicochemical property and can be measured more accurately. These and other factors can dictate whether a good predictive model can be built. 

Diverse Three TAACF datasets from PubChem 179 Naive Bayes, random forest, sequential minimal optimization, J48 decision tree. Used to create three models with different datasets. Naive Bayes had external test set accuracy 73-82.7, random forest 60.7-82.7%, SMO 55.9-83.3, and J48 61.3-80% Periwal et al. (36, 37)... 

Svetnik, V., Liaw, A., Tong, C., Culberson, J.C., Sheridan, R.P., and Feuston, B.P. (2003). Random forest A classification and regression tool for compound classification and QSAR modeling. Journal of Chemical Information and Computer Sciences, 43, 1947-1958. 

Figure 9. Random forest BBB predicted scores for molecules assigned as BBB+ and BBB — horizontal reference lines correspond to two decision thresholds. All predictions (scores) are for molecules not in the training set.

Figure 10. Model selection and assessment diagnostic performance measure S for random forest and partial least squares (PLS) methods applied to the BBB data for various percentages of the data (Ptrain) in the training set.

Breiman, L. (1999). Random forests, random features. Technical Report, University of California, Berkeley. 

Tumor classification by tissue microarray profiling random forest clustering applied to renal cell carcinoma. 

---