Skip to content

Instantly share code, notes, and snippets.

@gamazeps

gamazeps/pseudo.py

Created January 26, 2018 16:05
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save gamazeps/a894e35a1c51745f9d30fba6c51a2bc7 to your computer and use it in GitHub Desktop.
Save gamazeps/a894e35a1c51745f9d30fba6c51a2bc7 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
pseudo.py
def deepSv_predict(reads, reference, representation, model):
stats = generate_read_stats(reads) # we grab std and mean for insert and read size
candidates = paired_read_analysis(reads, stats) # generate candidate variants
calls = set()
for (variant, supporting_reads) in candidates:
proba = model.predict(representation.generate(variant, supporting_reads))
if proba > 0.5:
calls.add((variant, supporting_reads))
refined_calls = set()
for (variant, supporting_reads) in refined_calls:
located_call = split_read_location(reads, variant, reference)
if located_call is None:
located_call = supporting_reads_bounding(supporting_reads, variant)
refined_calls.add(located_call)
return format_to_vcf(refined_calls)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment