In 75% of the cases the means were larger than INCB018424 order the median (339 out of 450 subsets of measurements = 45 sets of measurements in 10 age groups). Logarithmic transformation reduced the number of cases to 42% (187), which is much closer to the expected percentage (50%). Two tests of normality were applied to each subset of the 450 measurements, the Kolmogorov–Smirnov test and the Shapiro–Wilks test. The original values returned 94 (21%) and 118 (26%) significant violations (α = 0.05) of the normality

assumption. The log-transformed values returned 33 (7%) and 42 (9%) violations, which is fairly close to the expected percentage (5%). The logarithmic transformation has the additional advantage that the estimated tolerance intervals do not include non-existing negative values. All values given in the tables are the re-transformed logarithmic values. To evaluate the age effect in the 45 sets of measurements a one-way ANOVA test (α = 0.05) was applied. The correlations of TACI and BAFF-R values with B cell subpopulations and age were assessed with the Pearson product-moment correlations and partial correlations. The logarithmic transformation was applied both to age in months (because of the large age range in the older groups; a value of 1 was added)

Apoptosis inhibitor and the measured values (because of their positive skewness). All calculations and tests were performed with spss 16.0 for Windows. Absolute B-lymphocyte numbers double during the first months of life and then gradually decrease almost fivefold from the second half of the first year of life to adult values; this is almost entirely caused by expansion of the naive B-lymphocyte pool, and to a small extent by expansion of transitional cells

(Fig. 1), which are higher in the youngest age groups. The absolute and relative sizes of the measured B-lymphocyte subpopulations are shown in Tables 1 and 2, respectively. The data were not normally distributed, given the means of the different subpopulations being larger than the median in 75% of the subsets of the measurements in the different age groups. We therefore used logarithmic values to calculate the value intervals (see ‘Material and methods’). With the provided reference values in Tables 1 and 2, we give a 95% chance that 90% of healthy children will show absolute numbers within this range. All sets of measurements showed a statistically significant Rucaparib age effect (α = 0.05), except for absolute and relative values of CD19+CD20- B cells; this subpopulation was very small in number in all age-groups. We determined TACI and BAFF-R expression in a randomly selected subgroup (total group n = 36; cord blood n = 6, 1 week to 2 months n = 2, 2–5 months n = 2, 5–9 months n = 3, 9–15 months n = 3, 15–24m n = 2, 2–5 years n = 2, 5–10 years n = 4, 10–16 years n = 4, adults n = 8). All children showed >95% BAFF-R positivity on CD19+ cells, with a mean fluorescence intensity of 226 (on a scale of 1024 channels).