Because this analysis was calculated on thresholds obtained before the intervention, data from both groups were used (n = 110, r = 0.744, p ≤ 0.001).
The cluster analysis was calculated on Euclidean distances between group centroids and the clustering procedure was UPGMA (Unweighted Pair Group Method with Arithmetic Mean, also called Paired group) using the PAST 3.11 software56.
Total error - including the true biological variation, as well as all sample (pre) processing procedures (islet isolation, beta cell FACS purification, protein extraction, tryptic digestion and LC-MS analysis) – was calculated on triplicate injections for all independent isolations (n = 3), after normalization.
The sample size for the primary analysis was calculated on the basis of a two-sided test for non-inferiority with a 5% significance level.
Clustering of related sequences based on amino acid homology was conducted using a Neighbour Joining (NJ) algorithm and bootstrap analysis was calculated on an unrooted NJ cladogram based on 1000 iterations using CLUSTALX [ 55].
CAM use between AA and CA is seen in Table 2. Chi square analysis was calculated on race and the demographic variables on CAM use, with significant findings for age (p =.003) and education (p <.001).
The discrimination analysis was calculated on the characters obtained from the ratios, except for the ratio of cone diameter to the number of cone scale rows (CD/CSR), because of the stable characteristic cone scale rows.
Duration of followup and survival in this analysis were calculated based on the date of the first day of salvage chemotherapy until the date of last contact, if the patient was still alive, or until the date of death.
Therefore, the MPR used in this analysis was calculated based on the assumption that participants used their own medications in hospital to reduce the risk of over-estimating adherence.
The power analysis was calculated based on the above findings from this exploratory study.
