tailor is a Tool for Adapter-domain Identification and
Linking Of RBPs.
install.packages("devtools")
devtools::install_github("sbthandras/tailor")# load package
library(tailor)
# load example data set
data(rbps)
# align first two RBPs
ps <- position_scores(
pattern = "MN395291-1",
subject = "ON513429-1",
id_var = "Core_ORF",
seq_var = "translation",
submat = "BLOSUM80",
data = rbps
)
ps$position_scores |> head(5)
#> # A tibble: 5 × 4
#> pattern subject identity score
#> <chr> <chr> <lgl> <int>
#> 1 M M TRUE 6
#> 2 N N TRUE 6
#> 3 I I TRUE 5
#> 4 L L TRUE 4
#> 5 R R TRUE 6
# find breakpoints along the alignment
bps <- find_breakpoints(ps)
bps
#> pattern_id subject_id start end mean_score pident
#> 1 MN395291-1 ON513429-1 1 152 5.138 0.947
#> 2 MN395291-1 ON513429-1 153 172 2.900 0.550
#> 3 MN395291-1 ON513429-1 173 630 -0.389 0.199
#> 4 MN395291-1 ON513429-1 631 645 -5.733 0.000
#> 5 MN395291-1 ON513429-1 646 913 -0.668 0.131
# identify the conserved N-terminal domain if it exists
adapter <- find_adapter(bps)
adapter
#> pattern_id subject_id start end mean_score pident
#> 1 MN395291-1 ON513429-1 1 172 4.878 0.901Note, for positions 1-152 and 153-172 the scores were quite different, but both ranges were classified as conserved, so they were merged (default). To investigate this, let’s plot the position scores:
plot(ps)
Depending on the method used for detecting breakpoints, the length of
the identified adapter may differ. By default, find_breakpoints() uses
method = "cemean" in which the function estimates the number of
breakpoints and their locations using the Cross-Entropy Method from the
breakpoint package. With method = "plateau" we get a different
result:
ps |> find_breakpoints(method = "plateau", type = "ewma") |> find_adapter()
#> pattern_id subject_id start end mean_score pident
#> 1 MN395291-1 ON513429-1 1 157 4.656 0.936Note that with this method we detect a shorter conserved N-terminal domain.
# load adapters of the example data set
data(adapters)
# convert to pident matrix and order by the original data frame
amat <- adapter_matrix(adapters, ids = rbps$Core_ORF)
# assign RBPs to clusters based on the distance matrix
clusters <- cluster_adapters(amat)
# visualise pident matrix with cluster assignments
plot(amat, clusters = clusters)
If you use tailor in a publication, please cite our paper:
András Asbóth, Tamás Stirling, Orsolya Méhi, Gábor Apjok, Victor Klein-Sousa, Nicholas MI Taylor, Hiba Hadj Mehdi, Balázs Papp, Eszter Ari, Bálint Kintses (2026) A global map of receptor-binding protein compatibility for the programmable design of Klebsiella and Acinetobacter phages. https://doi.org/10.64898/2026.02.20.706991
- Please report any issues, bugs or feature requests.
- License: MIT
- Get citation information for
tailorin R withcitation("tailor")