Calculate assembly factors from complex features reduce into a PCA plot.
Usage
X.predict.moonlighters(
data = NULL,
complexes = list(),
plot = TRUE,
sig.cutoff = 0.05,
moon.proportion.cutoff = 1/3,
min.reps = 1,
min.conditions = 1,
noPCs = 0,
weights = NULL,
kmeans.maxiter = 10,
kmeans.nstart = 5,
p.adj.method = "none"
)Arguments
- data
Data frame with features reduce by PCA. Columns are 'protein', and PC columns starting with 'PC'. Additional columns 'condition' and 'replicate' can be used.
- complexes
Named list: A list where each element is vector with proteins and the name of that element is the name of the protein complex.
- plot
Logical: Should the plots be plotted?
- sig.cutoff
Numeric: Significance level cutoff, 0.05 by default.
- moon.proportion.cutoff
Upper proportion of moonlighting subunits cutoff. Default is 1/3.
- min.reps
Integer: At least in how many replicates should a subunit be significantly moonlighting to appear in hits?
- min.conditions
Integer: At least in how many conditions should a subunit be significantly moonlighting to appear in hits?
- noPCs
Integer: How many principal components should be used for the clustering?
- weights
Numeric vector: weights for each principal component for k.means clustering. the complex centroid, distance from which is further calculated and gives
- kmeans.maxiter
Integer: Maximum interation for kmeans clustering.
- kmeans.nstart
Integer: How many starting sets should be tried for kmeans clustering?
- p.adj.method
Character string: One of the p.adj methods, see ?p.adjust. Default is "none".