In Gaura/cgr3d:

pos_a <- c( 0, 0)
pos_t <- c(10, 0)
pos_c <- c(10,10)
pos_g <- c( 0,10)

pos_start <- c(5,5)
pos_vertices <- list( a=pos_a, t=pos_t, c=pos_c, g=pos_g )

seq_of_interest <- "gaattc"
pos_trace <- matrix( rep( 0, 2 * (nchar(seq_of_interest)+1) ), ncol=2 )
pos_trace[1,] <- pos_start

for ( i in 2:nrow(pos_trace) ) {
  pos_trace[i,] <- ( pos_trace[i-1,] + pos_vertices[[substr(seq_of_interest, i-1, i-1)]] ) / 2
}

plot(pos_trace[,1],pos_trace[,2])

library(plotly)
aPos <- c(0,0,0)
tPos <- c(100, 0, 0)
cPos <- c(50,50*sqrt(3),0)
gPos <- c(50,25*sqrt(3),50*sqrt(3))
# aPos <- c(0,0,0)
# tPos <- c(100, 0, 0)
# cPos <- c(50,50*sqrt(3),0)
# gPos <- c(50,25*sqrt(3),0)
vertices <- list(a = aPos, t = tPos, g=gPos, c = cPos )
# pos_g <- c()
sequences <- Biostrings::readDNAStringSet( "repeats_dm.fa" )
sequence <- tolower(toString(sequences[1]))

# pos_trace <- matrix( rep( 0, 2 * (nchar(sequences)+1) ), ncol=3 )
# pos_start <- c()

# positions <- matrix(nrow = 4,ncol = 3)
# for(i in 1:3){
#       positions[1,i] <- aPos[i]
#       positions[2,i] <- tPos[i]
#       positions[3,i] <- gPos[i]
#       positions[4,i] <- cPos[i]
# }
split = 1/2
startPos <- c(50,25*sqrt(3),25*sqrt(3))
positions <- matrix(rep(0, nchar(sequence)), nrow = nchar(sequence)+1, ncol = 3)
for(i in 2:nrow(positions)){
      positions[i,]  <- (positions[i-1,] + vertices[[substr(sequence, i-1, i-1)]])*split
}

p <- plot_ly(type="scatter3d", x = positions[,1], y = positions[,2], z = positions[,3] )

p2 <- plot3dseq(tolower(toString(sequences[2])))

sequences <- Biostrings::readDNAStringSet( "repeats_sc.fa" )
p3 <- plot3dseq(getSequence(sequences,3))

p1 <- plot3dseq(getSequence(sequences,1))
p4 <- plot3dseq(getSequence(sequences,4))
p <- subplot(p1,p4)
p

sample( strsplit("AAATTTCCGG","")[[1]], 5000, replace=TRUE ) -> s
st <- paste0(s,collapse = "")

pr <- plot3dseq(tolower(st),1/2)
pr                  

# Load packages
library( "Biostrings")
library( "ggplot2" )
library( "cowplot" )
library( "readr" )
library( "stringr" )

# “Random” sequence
# paste0( sample( strsplit("AAATTTCCGG","")[[1]], 5000, replace=TRUE ), collapse="" ) -> random_seq
# random_seq <- tolower(random_seq)

# Starting conditions
pos_a <- c( 0, 0)
pos_t <- c(10, 0)
pos_c <- c( 0,10)
pos_g <- c(10,10)

pos_start <- c(5,5)
pos_vertices <- list( a=pos_a, t=pos_t, c=pos_c, g=pos_g )

# Load sequences
# repeats_dm <- readDNAStringSet( "repeats_dm.fa" )
# repeats_sc <- readDNAStringSet( "repeats_sc.fa" )

# Calculate trace and visualize
plot_cgr <- function( seq_of_interest, seq_name ) {
  pos_trace <- matrix( rep( 0, 2 * nchar(seq_of_interest) ), ncol=2 )
  pos_trace[1,] <- pos_start

  for ( i in 2:nrow(pos_trace) ) {
    pos_trace[i,] <- ( pos_trace[i-1,] + pos_vertices[[substr(seq_of_interest, i, i)]] ) / 2
  }

  df <- data.frame( pos_trace )
  df$pos <- 1:nrow(df)

  print(ggplot( df ) +
    geom_point( aes( X1, X2, color=pos ), size=1 ) +
    ggtitle( seq_name ))
  animation::ani.pause(0.01)
}

# list_seq   <- repeats_dm[c(1:6)]
# list_names <- names( list_seq )
# p <- mapply( function(x,y) plot_cgr( tolower(x), y ), list_seq, list_names, SIMPLIFY=FALSE )
# p2 <- plot_grid( p[[1]], p[[2]], p[[3]], p[[4]], p[[5]], p[[6]] )
# save_plot( "repeats_dm_1-6.pdf", p2, nrow=2, base_aspect_ratio=4 )

Biostrings::readDNAStringSet("col.fasta") -> collagen_seq
col_seq <- tolower(toString(collagen_seq[1]))
col_seq <- gsub('n','',col_seq)

readFasta <- function(file){
Biostrings::readDNAStringSet(file) -> collagen_seq
col_seq <- tolower(toString(collagen_seq[1]))
col_seq <- gsub('n','',col_seq)
}

plot_cgr(col_seq,"Collagen") -> col_plot

readFasta("dmd.fasta") -> dmd
dmd <- tolower(toString(dmdseq[1]))
plot_cgr(dmd,"DMD") -> pdmd

library("Biostrings")
Biostrings::readDNAStringSet("dmd.fasta") -> dmd
Biostrings::reverseComplement(dmd) -> rdmd
tolower(toString(rdmd)) -> rdmd
plot_cgr(rdmd,"Reverse DMD") -> prdmd

plot_animate <- function(seq_of_interest,seq_name){
      # Calculate trace and visualize
  pos_trace <- matrix( nrow = nchar(seq_of_interest), ncol=2 )
  pos_trace[1,] <- pos_start

  for ( i in 2:nrow(pos_trace) ) {
    pos_trace[i,] <- ( pos_trace[i-1,] + pos_vertices[[substr(seq_of_interest, i, i)]] ) / 2
  }

  df <- data.frame( pos_trace )
  df$pos <- 1:nrow(df)
  colnames(df) <- c('x','y')

  for(i in 2:nrow(df)){
        print(ggplot() + geom_point(aes(y = df$y[1:i], x = df$x[1:i]),size=1) + xlim(1,10) + ylim(1,10))
        animation::ani.pause(interval = 0.05)
  }

}

library("ggplot2")
fun <- function(seq_name){
for(i in 2:10){
      print(ggplot() + geom_point(aes(x=mdf$x[1:i],y = mdf$y[1:i]), size=1 ) + ggtitle( seq_name ))
      animation::ani.pause(interval = 0.2)
}
}

plot_cgr3d <- function( seq_of_interest, seq_name ) {
  pos_trace <- matrix( rep( 0, 3 * nchar(seq_of_interest) ), ncol=3 )
  pos_start <- c(0,0,0)
  pos_trace[1,] <- pos_start
  pos_trace[,3] <- c(1:floor(nchar(seq_of_interest)/2),ceiling(nchar(seq_of_interest)/2):1)
  for (i in 2:nrow(pos_trace) ) {
    pos_trace[i,] <- ( pos_trace[i-1,] + pos_vertices[[substr(seq_of_interest, i, i)]] ) / 2
  }
   p <- plot_ly(type="scatter3d", x = pos_trace[,1], y = pos_trace[,2], z = pos_trace[,3],
                   marker = (list (size = 0.5)))
}

Gaura/cgr3d documentation built on May 21, 2019, 11:07 a.m.