Figure S1

Derived shapes of the progression rate curve as a function of the current ADAS-cog score. Three structural models with different shape parameters were tested and the results of the assessment are shown below

Figure S2

Goodness-of-fit plots for the final model. (A) observed vs. population and individual predictions. The solid line represents the line of identity, (B) population weighted residuals vs. time and population predictions and (C) individual residuals vs. individual predictions, and distribution of population weighted residuals. Ordinate value of zero is presented in all the residual plots (solid line). Dashed line represents the LOWESS smoother. On the bottom right panel, the solid line represents the normal density and the dashed line represents the kernel density of population weighted residuals

Figure S3

Influence of ventricular volume on disease progression: (A) non-progressers without pathologic CSF [log CSF p-tau181P : Aβ1–42 ratio ≤ −1.86] and (B) progressers with pathologic CSF [log CSF /p-tau181P : Aβ1–42 ratio > −1.86]. Ventricular volumes were dichotomized to create roughly equal groups (> Median and ≤ Median) in the left and right panels where median ventricular volumes were 42.9 ml and 38.4 ml. Error bars represent standard error (SE) and lines represent simple linear regression through the data to allow visualization of the trends

Table S1

Elucidation of the motivation behind each stage of the modelling procedure

Table S2

Mixture model parameters for ADAS-cog scores from ADNI MCI subjects with CSF data

Table S3

Parameters of the mixture models fitted to the baseline CSF biomarker data in the ADNI MCI subjects. The MCI population was dichotomized based on the thresholds and the %CC statistic is reported using the post-hoc estimate of the sub-population assignment from the ADAS-cog mixture model

Table S4

Contingency table between CSF status and progresser status from the mixture model

bcp4308_sm_Data.doc149KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.