What does the scaled_mad mean?

Psy-Fer / SquiggleKit

SquiggleKit: A toolkit for manipulating nanopore signal data

MIT License

122 stars 23 forks source link

What does the scaled_mad mean? #20

Closed youyupei closed 5 years ago

youyupei commented 5 years ago

Hi James, May I ask a question? I noticed in your MotifSeq.py, when you doing the current normalization, why did you do this: scaled_mad = mad * 1.4826??

Psy-Fer commented 5 years ago

Hey!

Yep, read this https://en.wikipedia.org/wiki/Median_absolute_deviation

You will find how the constant scale factor is calculated assuming a normal distribution.

Let me know if that answers your question.

James

Psy-Fer commented 5 years ago

This is the relevant section

I hope this helps!

James

youyupei commented 5 years ago

Got it! Thanks, James, and sorry for the late reply. It is really helpful!

Psy-Fer commented 5 years ago

Wonderful. Glad I could help.