Mean error, biological data

A variety of statistical parameters have been reported in the QSAR literature to reflect the quality of the model. These measures give indications about how well the model fits existing data, i.e., they measure the explained variance of the target parameter y in the biological data. Some of the most common measures of regression are root mean squares error (rmse), standard error of estimates (s), and coefficient of determination (R2). 

Fig. 12.4 Functional coupling of the Gly389 and Arg389 receptors to adenylyl cyclase. (Reproduced from ref. 26 with permission of the American Society for Biochemistry and Molecular Biology.) Shown are the results from studies with clonal lines expressing each receptor at matched levels and the data presented as absolute activities (A) and normaUzed to the stimulation by forskolin (B). The results of similar studies with two other clonal lines are shown in C and D. The Arg389 demonstrated small increases in basal activities and marked increases in agonist-stimulated activities compared with the Gly389 receptor. Shown are the mean results from four independent experiments carried out with each line. Absent error bars denote that standard errors were smaller than the plotting symbol...

Figure 25 Sr concentrations in biological low-Mg calcite of brachiopods (dots), belemnites (crosses) and rudist bivalve (boxes). Mean values (bold curve) and two standard errors (thin curves) were calculated by moving a 20 Myr window in 5 Myr intervals across the data set. Ranges (vertical lines) and mean values (stars) of intrashell variations for concentrations in single rudist shells are also shown, but were not used in calculation of running means (source...

The data from the pilot survey(s) should be tested and analyzed to determine whether the error variation is homogeneous, normally distributed, and independent of the mean. If this is not the case (a situation which is highly probable with biological parameters), then the data could be appropriately transformed, or distribution-free (nonparametric) procedures could be used. 

All values presented for the chemical and biological analyses of plants are the means of five replicates. The mean and standard error were calculated for all data. PAH distributions in plants were one-way ANOVA tested. Statistical dififerences among individual treatments were assessed with Tukey-Kramer multiple comparison test at a significance level of a<0.05. All statistical analyses were performed with NCSS-2000, version 2001 (Copyright 2001 Jerry Hintze). 

---