Future warning when trying to find intersection between non overlapping ranges

[GH-MStamboulian]

m using pandas 1.5.0 & pyranges 0.0.127
this warning only arrises when there is no intersection between the two number intervals, for example the following code would trigger this warning:

import pyranges as pr

gr1 = pr.from_dict({"Chromosome": [1, 1], "Start": [1, 4], "End": [2, 27],}) 

gr2 = pr.from_dict({"Chromosome": [1, 1], "Start": [28, 35], "End": [40, 45]}) 
gr1.intersect(gr2)

The part of the implementation that is triggering this warning (inside the pyranges package) is the following segment:

pd.options.mode.chained_assignment = None  # default='warn'
scdf.loc[:, "Start"] = new_starts
scdf.loc[:, "End"] = new_ends
pd.options.mode.chained_assignment = "warn"

I could suppress this warning however im wondering if I should actually not be ignoring it or if the codes behavior might change at some point in the future.

when I calculate the overlap between the same two ranges I do not get this warning.

gr1.overlap(gr2)

can you please help me understand what this warning message would be and if there is a risk for it to mutate any of my ranges gr1 or gr2 in future? Thank you!

[GH-MStamboulian]

@marco-mariotti

[endrebak]

I can have a look on Monday. Thanks.

[endrebak]

There is no risk here that the object passed in will be changed.

reindex creates new objects:

scdf, ocdf = scdf.reindex(_self_indexes), ocdf.reindex(_other_indexes)

[GH-MStamboulian]

@endrebak thank you for looking into this. So what would be the most appropriate way to get rid of this warning? should I just suppress it?

[endrebak]

I am a bit surprised that it shows up. If you look at the code, you see that I turn off the warning.

[GH-MStamboulian]

yes I have no idea why its still complaining. what pandas version are you using?

[endrebak]

2.0.1

[GH-MStamboulian]

@endrebak I see, and do you still get this warning too with the newer pandas version?

[cmatKhan]

I believe I am seeing this, also, though I'm not sure it is the same FutureWarning.

test_gtf.txt

x = pr.read_gtf('test_gtf.txt')

x.features.introns(by='transcript').intersect(x[x.Feature=='exon'],strandedness='same', how='containment')

result:

	Chromosome	Source	Feature	Start	End	Score	Strand	Frame	gene_id	transcript_id
0	chr1	unknown	intron	15	19	.	+	.	gene1	transcript_1
1	chr1	unknown	intron	15	19	.	+	.	gene1	transcript_1

FutureWarning: The `inplace` parameter in pandas.Categorical.add_categories is deprecated and will be removed in a future version. Removing unused categories will always return a new Categorical object.
  introns.Feature.cat.add_categories(["intron"], inplace=True)

FutureWarning: In a future version, `df.iloc[:, i] = newvals` will attempt to set the values inplace instead of always setting a new array. To retain the old behavior, use either `df[df.columns[i]] = newvals` or, if columns are non-unique, `df.isetitem(i, newvals)`
  introns.loc[:, "Feature"] = "intron"

pyranges version info:

{'pyranges version': '0.0.120', 'pandas version': '1.5.2', 'numpy version': '1.23.0', 'python version': sys.version_info(major=3, minor=9, micro=18, releaselevel='final', serial=0), 'ncls': '0.0.65', 'sorted_nearest': '0.0.37', 'pyrle': '0.0.35', 'ray': 'not installed', 'bamread': 'not installed', 'pyranges_db': 'not installed', 'pybigwig': 'not installed', 'hypothesis': 'not installed'}