The Multi-Ontology Enrichment Tool (MOET) Algorithm

Calculation of p-value *New MOET v.2 formula MOET calculates the p-value of the over-represented terms annotated to the genes within the input gene list based on the Hypergeometric distribution test with a lower p-value indicating greater over-representation. The formula for calculating P or probability of seeing at least k genes associated with a given ontology term is calculated as follows:                    (M choose k) (N-M choose n-k) P(X=k) =   —————————————————-                          (N choose n) Further, p-value is calculated as: p-value = P(X=k)+P(X=k+1)+……P(X=n) (Sum of probabilities from k=input annotated genes to k=total input genes ) Where; X = a random variable n = size of the input gene list N = total number of genes in the reference population k = number of genes annotated to a ontology term in the input list M = number of genes annotated to the ontology term in the reference population. The reference population for the Hypergeometric test in MOET includes all the genes in the selected species annotated for that term. Bonferroni Multiple Test Correction Bonferroni Multiple Test Correction is used to reduce the probability of false-positive (Type I) errors resulting from multiple statistical tests (for each ontology term) being run at the same time. The adjusted significance level (p-value) is calculated as the nominal (or uncorrected original) p-values multiplied by the number of independent tests to yield the working p-values. Adjusted p-value = p-value x (number of tests) Odds Ratio Calculation *Additional feature in MOET v.2 Formula to calculate odds ratio = (a*d) / (b*c) a = annotated input gene count = k c = annotated ref gene count – a = M – k b = input gene count – a = n – k d = ref gene count – annotated ref gene count – b = N – M – n + k