Although codon usage bias has been identified and studied in various species, its evolutionary effects remain largely unknown. Furthermore, codon usage bias has rarely been used in algorithmic design or as a character state in phylogenetic tree reconstruction, meaning its use as an algorithmically informative character largely remains untested.
The development and implementation of accurate biological algorithms has become paramount to biological research as the amount of data biologists generate exceeds the analysis capabilities of traditional algorithms and techniques. While a plethora of algorithms currently exist to identify orthologs, many lack sufficient documentation to easily use the software. Furthermore, many algorithms were initially developed to address much smaller problems for smaller data sets, meaning they are computationally intensive in both time and memory. Some algorithms perform poorly on certain data sets, and it is often unclear what limitations exist for a given algorithm. Algorithmic development, therefore, is necessary to address the issues of computational complexity; however, usability is also necessary to ensure that biologists with limited computational background understand an algorithm’s use and limitations. In the process of developing fast, user-friendly algorithms, I hope to uncover algorithmically unutilized genetic biases to identify orthologs faster and more accurately.