Plot GC content from user provided dataframe (.tsv)

[LFelipe-B]

Dear moshi4, wish to thank you for the earlier implementation of using gff multiple contig file as input from the user, it was really helpful. Can I ask now for assistance on how to plot GC content from an uploaded table (or other type of similar information) as colorbar (or lines for sliding window for ie. as a separate track? The examples only show the calculations from .gbk files and I don't have these type of file, only .gffs. I think it would be useful for the whole community, since your tool is great and really user friendly!

I assume that here in the colorbar example: https://moshi4.github.io/pyCirclize/plot_tips/ user could provide the dataframe in the section matrix1 = np.random.randint(vmin1, vmax1, (5, 100)) ?

And for the lines I am not sure. Another idea that I had was to provide the GC value just for the CDS and map the prot_ID to the specific value, but no clue on how add this data to a specific track. Hope I am clear, and sorry if this is a really trivial problem for python users. Thank you!

[moshi4]

Here is an example implementation code for your question below.

• How to plot GC content from a file
• How to plot GC content with fill_between, heatmap(with colorbar)

from pycirclize import Circos
import pandas
from pycirclize.parser import Genbank, Gff
from pycirclize.utils import load_prokaryote_example_file

# Generate GC content example file (gc_content.tsv)
gbk_file = load_prokaryote_example_file("enterobacteria_phage.gbk")
gbk = Genbank(gbk_file)
pos_list, gc_contents = gbk.calc_gc_content()
gc_content_df = pandas.DataFrame({"Position": pos_list, "GCcontent": gc_contents})
gc_content_tsv_file = "gc_content.tsv"
gc_content_df.to_csv(gc_content_tsv_file, sep="\t", index=False)

# Plot genomic features & GC content (from GFF & GC content file)
gff_file = load_prokaryote_example_file("enterobacteria_phage.gff")
gff = Gff(gff_file)

circos = Circos(sectors={gff.name: gff.range_size}, start=0, end=320)
sector = circos.sectors[0]

outer_track = sector.add_track(r_lim=(100, 100))
outer_track.xticks_by_interval(5000, label_formatter=lambda v: f"{v / 1000:.0f} Kb", show_bottom_line=True)
f_cds_track = sector.add_track(r_lim=(94, 98), r_pad_ratio=0.1)
f_cds_track.genomic_features(gff.extract_features(target_strand=1), fc="tomato")
r_cds_track = sector.add_track(r_lim=(90, 94), r_pad_ratio=0.1)
r_cds_track.genomic_features(gff.extract_features(target_strand=-1), fc="skyblue")

df = pandas.read_csv(gc_content_tsv_file, sep="\t")
pos_list, gc_contents = df["Position"].to_numpy(), df["GCcontent"].to_numpy()
max_gc, min_gc = max(gc_contents), min(gc_contents)

gc_fill_track = sector.add_track(r_lim=(75, 85))
gc_fill_track.grid()
gc_fill_track.fill_between(pos_list, gc_contents, y2=min_gc, vmin=min_gc, color="lightgrey")
yticks = [min_gc, max_gc]
yticks_label = [f"{y:.1f}" for y in yticks]
gc_fill_track.yticks(yticks, yticks_label, vmin=min_gc, label_size=6, side="left")

gc_heatmap_track = sector.add_track(r_lim=(65, 70))
gc_heatmap_track.heatmap(gc_contents.reshape(1, -1), cmap="bwr")
circos.colorbar((0.3, 0.5, 0.4, 0.02), vmin=min_gc, vmax=max_gc, cmap="bwr", orientation="horizontal", colorbar_kws=dict(label="GC Content (%)"))

circos.text("Forward CDS ", r=96, ha="right", size=8)
circos.text("Reverse CDS ", r=92, ha="right", size=8)
circos.text("GC content (fill_between) ", r=80, ha="right", size=8)
circos.text("GC content (heatmap) ", r=67.5, ha="right", size=8)

circos.savefig("gc_content_plot_example.png", dpi=300)

gc_content.tsv

Position	GCcontent
0	23.333333333333332
60	38.333333333333336
120	53.333333333333336
180	52.5
...
60840	35.0
60900	33.33333333333333
60942	33.33333333333333

gc_content_plot_example.png

With these code examples, I think you can achieve what you want to do.
I do not intend to explain and teach everything in detail, so please do your best with this code examples.

[LFelipe-B]

Thank you again moshi4, you were really helpful! Cheers