How does the Fluidigm SNP Genotyping Analysis software determine the genotype call for a sample?
Allele discrimination is primarily based on ratios from the two normalized probe measurements.
The Fluidigm SNP Genotyping Analysis software uses the k-means clustering analysis method to automatically call genotypes. The algorithm starts by initializing a number of cluster centers (k), first assuming all three clusters are present (k=3). It is followed by an iterative process of assigning datapoints to clusters based on the proximity to the cluster center and updating cluster centers based on the means of the datapoints within each cluster.
If the algorithm fails to identify three clusters, then a k of two clusters is tried. Clusters must be sufficiently far apart from each other. If they are not, clusters may receive the same genotype assignments.
A review of automatic genotype calls in individual scatter plots is recommended after cluster analysis.