plotDbOverlap - Plot heatmap of one or two db features overlap

Description¶

Plot a matrix to visualize the the number of objects (e.g. clones and sequences) shared between groups.

Usage¶

plotDbOverlap(
db,
group = "sample",
features = c("clone_id", "sequence_alignment"),
heatmap_colors = c("white", "orange", "grey80"),
print_zero = FALSE,
long_x_angle = 90,
title = NULL,
xlab = NULL,
ylab = NULL,
plot_order = NULL,
silent = F,
similarity = c("min", "jaccard"),
na.rm = FALSE,
identity = c("exact", "ambiguous", "ham_nt", "ham_aa"),
threshold = 0,
geom_text_size = 3
)

Arguments¶

db

Changeo db data.frame

group

Vector with column names for grouping. Overlap will be calculated across the groups. e.g SAMPLE

features

Column name of the feature column(s) shared across group. e.g. CLONE, SEQUENCE_INPUT

heatmap_colors

Vector of colors representing low and high values on the heatmap and the diagonal. Default is c(“white”,”orange”, “grey80”)

print_zero

Show labels on zero overlap cells or not.

long_x_angle

Angle to rotate x axis labels when any of the labels is longer than 6 characters

title

A string that will be used for the heatmap title. If NULL, group and featureS will be used.

xlab

Text to be used as x axis title

ylab

Text to be used as y axis title

plot_order

A vector to reorder the grouped columns. Can contain either the specific names of the grouped columns (e.g. c("Donor1","Donor2") ); the names of columns in db (e.g. "DONOR", which has factor levels “Donor1” and “Donor2” or NULL if ordering is not relevant.

silent

If T, the plot will not be printed, it will be found in the returned list.

similarity

vector of the same length as features. For each feature, method used to quantify the overlap. “min” will use the number of shared features over the number of features in the smaller set. “jaccard” will use the jaccard index, expressed as a percent, and defined as the intersection of features over the union. Can also be one of the possible values in the last grouping column. For example to make the percent always relative to the memory samples, use group="SAMPLE", "SORT") and similarity=c("memory")

na.rm

logical. If TRUE, NA values will be removed and not considered

identity

Vector of the same length as features spcifying how to establish identity. For each features, compare the exact value of the features (identity='exact'), allow ambiguous characters in DNA sequences (identity='ambiguous') using the function seqEqual, or use hamming distance and a threshold (identity='ham'). See details.

threshold

Identity threshold to be used when identity='ham'.

geom_text_size

Plot text size

Value¶

A list with the plot object and a data.frame with the values.

Details¶

Can be used to visualize the number of clones and/or sequences shared between compartments, which are potentially antigen-specific

Note: When using exact=FALSE and distance="min" the overlap will be calculated using the number of shared features from the smallest set. In case of ties, the smallest number of shared features.

Examples¶

### Not run:
data("ExampleDb", package="alakazam")
# db <- ExampleDb
# 
# ## Plot the number of sequences that overlap across samples
# overlap <- plotDbOverlap(db, group="sample_id", features="clone_id", identity="exact", similarity="jaccard")
# overlap <- plotDbOverlap(db, group="sample_id", features=c("clone_id","junction"))
# 
# ## The returned plot can be modified
# ## To edit the axis labels. the title and change the color scale and change
# ## the theme
# overlap$p + xlab("Sequence") + ylab("Clone") + ggtitle("New title") +
#  scale_fill_gradient(low="white", high="orange", na.value="black") + theme_enchantr()