Frequency of Preferred Codons (FOP)
Determine which codons are preferred in a set of highly expressed genes, and then
calculate the frequency of preferred codons (pref/(pref+nonpref)) for each gene.
Major Codon Usage (MCU)
Use multivariate analysis to first identify the axis of greatest variance in 59-dimensional
codon usage frequency data (RSCU), then identify which codons contribute positively to this
trend.
Codon Adaptation Index (CAI)
Determine degree to which codons are preferred in a set of highly expressed genes
(weight = count of a given codon/count of its maximal sibling), and then
calculate the adaptiveness for each gene by taking the geometric mean for the
weights associated with the codons in each gene.
Self-Consistent Codon Index (SCCI)
Iteratively searches for a self-consistent set of genes to use as a reference set, then
calculate CAI/SCCI as in above (CAI).
Effective Number of Codons (Nc)
Statistically determined metric of codon preference. If a single codon is utilized
(to the exclusion of all others) when coding for each amino acid then the effective number
of codons would be 20 (there are 20 amino acids). If there is balanced codon usage
(no bias or preference) then the effective number of codons would be 59 (there are 59 codons
that code for amino acids that can be coded for by more than one synonymous codon).
Scaled Chi Squared (Χ2)
A chi squared calculation is performed that examines deviation of codon usage from expected
values.
Transfer RNA Adaptation Index (tAI)
Similar to CAI, but the weights are determined by examining tRNA gene copy number for the
associated codons.
Modified Self-Consistent Codon Index (mSCCI)
Similar to SCCI but it directs the search for a reference set away from unbalanced GC-content
genes and toward the genes that are more likely to exhibit translational efficiency bias.
Direct Search Approach Using a Genetic Algorithm
Instead of acquiring a set of known highly expressed genes and tallying codons to determine
codon adaptiveness, this meathod performs a search for a set of weights that explains the
high placement of known highly expressed genes in a sorted list of all genes (by CAI score).
tRNA Gene Data
This is not a bias detection method. The linked article presents methods for identifying tRNA genes.
This information is useful (particularly for determining the gene copy number) in methods like tAI.