How to find which coordinates are repeating in target chromosomes

[khush876]

Hi,

It is very useful tool and the graphics are amazing. However, I have some issues if you can address

1. How to add color to the synteny analysis as black and white is not appropriate
2. Is there a way I can plot (plot_synteny) of all the chromosome as a one figure
3. Plot_coverage gives very nice graphics, but I am intrested to know which segments of one chromosome is recombinant in other. I have 9 chromosome (target) and it seems that my query is showing lot of recombination. Is there a way to find out more detail (or circular plot) to map those synteny like BioCircos?
   Thank you

[dwinter]

Hi @khush876 ,

Sorry for the delay in responding to this. I have recently started a new position, so been a bit hectic finishing up one job and startng the next.

Some answers to you points,

(1) This is something we can add quite easily, for now you could use this function in place of plot_synteny

plot_synteny_colourfullly <- function (ali, q_chrom, t_chrom, centre = TRUE, rc = FALSE, xlab = "Position in query", 
    ylab = "", x_labeller = Mb_lab, ribbon_colour="grey80", seq_colour="white") 
{
    synt_df <- pafr:::synteny_data(ali, q_chrom = q_chrom, t_chrom = t_chrom, 
        rc = rc)
    cs <- pafr::chrom_sizes(ali)
    tlen <- cs$tlens$tlen[cs$tlens$tname == t_chrom]
    qlen <- cs$qlens$qlen[cs$qlens$qname == q_chrom]
    seq_lens <- data.frame(seq = factor(c(q_chrom, t_chrom)), 
        start = 0, end = c(qlen, tlen))
    if (centre) {
        os <- abs(tlen - qlen)/2
        if (qlen > tlen) {
            synt_df[["x"]][synt_df[["seq"]] == t_chrom] <- synt_df[["x"]][synt_df[["seq"]] == 
                t_chrom] + os
            seq_lens[2, 2:3] <- seq_lens[2, 2:3] + os
        }
        else {
            synt_df[["x"]][synt_df[["seq"]] == q_chrom] <- synt_df[["x"]][synt_df[["seq"]] == 
                q_chrom] + os
            seq_lens[1, 2:3] <- seq_lens[1, 2:3] + os
        }
    }
    if (q_chrom > t_chrom) {
        bottoms <- c(2.05, 0.8)
        tops <- c(2.2, 0.95)
    }
    else {
        bottoms <- c(0.8, 2.05)
        tops <- c(0.95, 2.2)
    }
    ggplot() + geom_polygon(data = synt_df, aes_string(x = "x", 
        y = "seq", group = "block_id"), fill = ribbon_colour, colour = "black") + 
        geom_rect(data = seq_lens, aes_string(xmin = "start", 
            xmax = "end", ymin = "bottoms", ymax = "tops"), colour = "black", 
            fill = seq_colour) + scale_x_continuous(xlab, labels = x_labeller) + 
        ylab(ylab)
}

Using the example from the README you could do

#beautiful....
plot_synteny_colourfullly(long_ali, q_chrom="Q_chr3", t_chrom="T_chr4", centre=TRUE, seq_colour = "forestgreen", ribbon_colour = "pink")

(2) In short no. I think doing this well would be quite difficult (how to order target and query sequences, etc) so it's not something that is likely to be on the horizon. I have produce nice figures for thiese sorts of plots with MCScan (https://github.com/tanghaibao/mcscan) and if we get #26 of the ground we might be able to make whole genome circos plots with similar effects

(3)I'm not sure exactly what you are after, with #26 we hope to make it quick and easy to produce circos plots. It is also possible to extract the location of a given aligned-block from the .paf table, and the next release will allow you to extract bed format corrdinates from an alignment. If you have a specific you don't mind sharing you can send it to me at [email protected] and I can try to make some time to look at it?

[khush876]

Hi David,

Thank you so much for sending the code. It is really great. It gives really beautiful synteny plot. I am very happy that I can make different color for each chromosome.

2. In this question, I was talking about plotting, all q_chrom and t_chrom from .paf file into one plot

plot 1 <- q_chrom="Q_chr1", t_chrom="T_chr1"
plot 2 <- q_chrom="Q_chr2", t_chrom="T_chr2"
plot 3 <- q_chrom="Q_chr3", t_chrom="T_chr3"

superplot <- plot 1, plot 2, plot 3

3. I will send you my plots on your email.

Thank you