An R script to measure the potenial influence of different paramaters which could have effected the speed of pairwise sequence alignment, using the "pairwsieAlignment" function of the Biostrings package.
Make sure to have installed the following packages : Biostrings, dplyr, ggplot2, and here
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("Biostrings")
install.packages("dplyr")
install.packages("ggplot2")
install.packages("here")
The Script uses the functionalities provided in Base-R (system.time() command) to measure how long it takes for the computer to run a pairwise alignment task. I have first generated random sequences of DNA and Protein to use as data, then performed alignments. One might assume that it should take different times for the system to perform the task depending that the requested alignment is for Protein or DNA, for Local or Global alignment, with short sequneces or long sequences. For each setting of the test (AA/DNA, Local/Global), there are 5 replicate random sequences. See the "SAST_100.csv" file in the Expected_Results directory.
The script records all the times in seconds, and finally outputs the results in a csv file.
The NSEQ variable determines how long the random sequences should be for the speed test.
NOTE: The result "SAST_100.csv" shows that the type of alignment(Local or Global) or the nature of the sequences (AA or DNA) DOES NOT affect the time needed to complete the alignment, but only the size of the sequence plays a critical role.
NOTE: I haven't performed the statistical analysis and visualization on the dataset yet, maybe in the future.
NOTE: I just found out that the choice of substitutionMatrix in the pairwiseAlignment function, greatly changes the times of processing. The script uses the NULL option (default), however I might change it to DNAFULL for DNA and BLOSUM62 for AA in the future.
NOTE: As always, I have put the Expected_Results folder which contains the files that the R script generates once you run it.
NOTE: Make sure to edit setwd command in the R script to point at your desired directory. In this particular repository, it is not necessary because of the here package. However you can consider to do it manually in case the script somehow does not work for you.