Predicting the impact of mutations on protein function

===============================================================

Installation

To install siftr, first make sure your R version is at least R 3.0. You can check by typing the following into your R console:

R.Version()$major

Install and load devtools package:

install.packages("devtools")
library("devtools")

Download and install the siftr package from github:

install_github("omarwagih/siftr")

Load the siftr package

library("siftr")

Running siftr on sample data

# Get the path to the sample amino acid alignment
# Alternatively you can pass in the alignment as a vector
# e.g. c('SSSS', 'STTT', 'SYSY', 'ASST', 'KKHS')
sample_fa = system.file("extdata", "P39709.alignedfasta", package = "siftr")


# Compute sift scores
# You can adjust the number of cores for faster processing
sift_mat = predictFromAlignment(sample_fa, cores=1)

# Summary of the sift matrix
summary(sift_mat)

The generated matrix contains the sift scores. Each row is the substituted amino acid and each column is the position in the alignment.

Then, filter the matrix for significant scores (< 0.05), and output in a table format:

# Keep predictions with sift scores < 0.05 
filt = filterPredictions(sift_mat, score_thresh = 0.05)

# Display resulting table
print( head(filt) )

You can extract further information about the alignment from functions diffAA and infoContent

# Get information content per position for the alignment
info_content = infoContent(sift_mat)

# Get number of unique amino acids per position
diff_aa = diffAA(sift_mat)

That's all there is to it!

For further documentation see ?predictFromAlignment, ?filterPredictions, ?diffAA and ?infoContent

Contact

If you have any feedback, suggestions or questions, please drop me a line at (wagih(at)ebi.ac.uk) or open an issue on github.