search

mwaskom / lyman

Data pipelines and analysis library for functional MRI
http://www.cns.nyu.edu/~mwaskom/software/lyman
BSD 3-Clause "New" or "Revised" License
54 stars 22 forks source link

Develop nuisance components of linear model #188

Closed mwaskom closed 5 years ago

mwaskom commented 5 years ago

This PR mainly adds nuisance variables to the main Lyman model.

Nuisance signals can be estimates from four sources (WM and CSF from the aseg, an "edge of brain" mask, and the voxels defined as locally noise during preprocessing) using a PCA dimensionality reduction over voxels.

It may be advisable in a future PR to refactor the modeling code, getting access to the nuisance variables outside of the context of the workflows. Currently (and have a fair bit of code repetition).

A few other small issues are rolled into this PR (see commits).

codecov-io commented 5 years ago

Codecov Report

Merging #188 into master will not change coverage. The diff coverage is 100%.

Impacted file tree graph

@@          Coverage Diff           @@
##           master   #188    +/-   ##
======================================
  Coverage     100%   100%            
======================================
  Files          10     10            
  Lines        2189   2302   +113     
======================================
+ Hits         2189   2302   +113
Impacted Files Coverage Δ
lyman/glm.py 100% <100%> (ø) :arrow_up:
lyman/signals.py 100% <100%> (ø) :arrow_up:
lyman/workflows/model.py 100% <100%> (ø) :arrow_up:
lyman/utils.py 100% <100%> (ø) :arrow_up:
lyman/workflows/preproc.py 100% <100%> (ø) :arrow_up:
lyman/visualizations.py 100% <100%> (ø) :arrow_up:
lyman/frontend.py 100% <100%> (ø) :arrow_up:

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update db1566e...d049c82. Read the comment docs.