plot-methods {procoil} | R Documentation |
Functions for plotting prediction profiles
## S4 method for signature 'CCProfile,missing' plot(x, col=c("red", "blue"), standardize=TRUE, shades=NULL, legend="default", legendPos="topright", xlab="", ylab="weight", lwd.profile=1, lwd.axis=1, las=1, heptads=TRUE, annotate=TRUE, ...)
x |
Object of class |
col |
Character string containing the name(s) of the color(s) in which the profile(s) should be plotted. |
standardize |
If |
shades |
Vector of at least two color specifications (default:
NULL). If not NULL, the background area above and below the base
line y=-b/L are shaded in colors |
legend |
A character string containing the legend/description of
the profile. If |
legendPos |
position specification for legend (if |
xlab |
label of horizontal axis, empty by default. |
ylab |
label of vertical axis, defaults to “weight”. |
lwd.profile |
profile line width as described for
parameter |
lwd.axis |
axis line width as described for
parameter |
las |
see |
heptads |
if |
annotate |
if |
... |
all other arguments are passed to the
|
The plot
function displays a prediction profile as a step
function over the sequence with the steps connected by vertical lines.
The sequence and the heptad register are visualized below and above
the profile, respectively. The baseline value -b/L and the light
gray line has the following meaning: It is obvious that we can rewrite
f(x)=b+sum over all si(x) for i=1,… L
as
f(x)=sum over all (si(x) - (-b/L)) for i=1,… L,
so the discriminant function value f(x) can be understood as the sum of values (si(x) - (-b/L)), i.e. the area between the constant value -b/L and the prediction profile. If the area above the light gray line is greater than the area below the light gray line, the sequence is predicted as trimer, otherwise as dimer.
If plot
is called for a CCProfile
object
that contains profiles of two sequences, the two profiles are plotted
together to facilitate a comparison of profiles (e.g. wild type
sequences versus mutants). Although the plot
function tolerates
profiles/sequences with different lengths and/or unaligned heptad
registers, it is obvious that the superimposition of profiles of
two unaligned, unrelated sequences makes little sense.
The plot
functions gives an error if is called for a
CCProfile
object that contains profiles of
three or more sequences.
The given function is only a wrapper around the
plot
function provided
by the kebabs package. The only difference is that heptad
seperators (argument heptads
) and the heptad annotation
(argument annotate
) are displayed by default.
Moreover, presently, no legend is displayed by default if a
single profile is plotted for an unnamed sequence.
This function does not return any value.
Ulrich Bodenhofer bodenhofer@bioinf.jku.at
http://www.bioinf.jku.at/software/procoil/
Mahrenholz, C.C., Abfalter, I.G., Bodenhofer, U., Volkmer, R., and Hochreiter, S. (2011) Complex networks govern coiled coil oligomerization - predicting and profiling by means of a machine learning approach. Mol. Cell. Proteomics 10(5):M110.004994. DOI: 10.1074/mcp.M110.004994
Palme, J., Hochreiter, S., and Bodenhofer, U. (2015) KeBABS: an R package for kernel-based analysis of biological sequences. Bioinformatics 31(15):2574-2576. DOI: 10.1093/bioinformatics/btv176
## predict oligomerization of GCN4 wildtype GCN4wt <- predict(PrOCoilModel, "MKQLEDKVEELLSKNYHLENEVARLKKLV", "abcdefgabcdefgabcdefgabcdefga") ## plot profile plot(GCN4wt) ## define two GCN4 mutations GCN4mSeq <- c("GCN4wt" ="MKQLEDKVEELLSKNYHLENEVARLKKLV", "GCN4_N16I_L19N"="MKQLEDKVEELLSKIYHNENEVARLKKLV") GCN4mReg <- rep("abcdefgabcdefgabcdefgabcdefga", 2) ## predict oligomerization GCN4mut <- predict(PrOCoilModel, GCN4mSeq, GCN4mReg) ## overlay plot of the two profiles plot(GCN4mut)