Databases comparative

With this database in hand, a simple question is asked [29] How different is a knowledge-based potential derived from this lattice database compared to the actual energy function used to construct the database If statistical errors are negligible and the knowledge-based method is perfect, the answer is expected to be They are exactly the same. ... 

In order to be appUed the models should be predictive. Unfortunately, the models frequently fail and demonstrate significantly lower prediction ability compared to the estimated one, when they are applied to new unseen data [100-103, 106]. One of the main reasons for such failures can be the lack of available experimental data and difficulties in calculating log D, as discussed in Section 16.4.2. Another problem of low prediction ability of log D models can be attributed to different chemical diversity of molecules in the in-house databases compared to the training sets used to develop the programs. 

The large number of compounds enables satisfactory description of the retention behavior of the target analyte within the limited range of chromatographic conditions. The greater relevance of the compounds in the database compared to target analyte reduces the effects of unmodeled phenomena, since any compound that is predicted will have the most similar compounds... 

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

Figure 15.10 shows the ratio of permeability coefficients for the 14 compounds common to the snake and human databases compared with predictions plotted as a function of log and MW. The two steroids deoxycorticosterone (DC, called deoxycorticosterone in tire animal investigation and eortexone in the human investigation)... 

Figure 15.9 Ratios of average permeability coefficients for 31 compounds common to the validated hairless mouse and human databases compared to the average ratio of 3.1 (solid horizontal line) and to the difference behveen the regressions developed from the hairless mouse database and the fully validated human databases (a) plotted as a function of log K tor MW= 100 (short dashes) and MW = 300 (long dashes) (b) plotted as a function of MW for log K =2 (short dashes) and log = 4 (long dashes).

The precipitate hardening efficiency iOyp/fp) vs fp for solute cluster hardening within the IVAR database, compared with trends derived from the Russell-Brown (RB) and Bacon and Osetsky (B-O) models. o ,p, irradiation-induced yield stress increment associated with precipitation fp, volume fraction of precipitates fp, average precipitate radius derived from SANS measurements. 

Different problems for substructure searching on chiral compounds arise in two-dimensional structure databases compared with three-dimensional databases. The two-dimensional case is the worse of the two, because two-dimensional databases... 

The importance of the database used in parametrizing the MM function is revealed by the historical development of the field. In early studies, it was natural and appropriate to use experimental data (structures, thermodynamic properties, etc.) to optimize the MM parameters. As ab initio calculations have become more accurate, it is being realized that the results of such calculations provide a larger, more complete, and much more self-consistent database compared to experimental data, and parametrization of MM functions has depended increasingly on utilizing ab initio properties. 

Figure 5 Descriptor histograms and SE and SSE values from two different compound databases. Compared are distributions of descriptor values in a chemical (ACD) and pharmaceutical (MDDR) database. The top number in the upper part of each chart reports the SE value, and the number beneath is the SSE value. Descriptor abbreviations MW, molecular weight a don, number of hydrogen bond donor atoms b rotN, number of rotatable bonds in a molecule logP(o/w), logarithm of the octanol/water partition coefficient.

---