Why it is "p≤0.05" but not exact number?

[huajiasha]

test code below

    pairs = [
        ("PBS", "OVA")
    ]
    annot = Annotator(ax2, pairs,fontsize=10, **plotting_parameters)
    annot.configure(test='Mann-Whitney', text_format='simple', loc='inside',show_test_name=False, verbose=2)
    annot.apply_test()

the p value is 1.476e-02 or 0.01476, it is far from 0.05

<ipython-input-20-bed6cdf16d56>:8: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

  ax2=sns.boxplot(ax=ax,palette='Set2',width=0.5,**plotting_parameters)
PBS vs. OVA: Mann-Whitney-Wilcoxon test two-sided, P_val:1.476e-02 U_stat=9.000e+00

[johnmbarrett]

Not affiliated with the statannotations developers in any way, just ran into something similar myself just now.

This appears to be the intended behaviour of the simple text format, i.e. it prints p less than various thresholds rather than the actual p values. If two decimal places is enough, you can pass pvalue_thresholds=[] to annot.configure(). If you want more control, you can pass text_format='full',pvalue_format_string='{:3.1e}' (or whatever format you actually want) instead.

[huajiasha]

Not affiliated with the statannotations developers in any way, just ran into something similar myself just now.

This appears to be the intended behaviour of the simple text format, i.e. it prints p less than various thresholds rather than the actual p values. If two decimal places is enough, you can pass pvalue_thresholds=[] to annot.configure(). If you want more control, you can pass text_format='full',pvalue_format_string='{:3.1e}' (or whatever format you actually want) instead.

thanks for sharing, but some p value show exact numbers not just p<=0.05, p<0.01, p<0.001, etc.
So this behaviour of the simple text format, is not consistent.
I will try text_format='full',pvalue_format_string='{:3.1e}'