Documentation of Anatomy Analysis Code

This is the analysis code for the striatal portion of my PhD thesis work: "A comprehensive map of excitatory input convergence in the mouse striatum," (Publication in Review). The corticostriatal dataset was generated from data produced by the Allen Institute for Brain Science (AIBS) & the thalamostriatal dataset was generated from data produced by us for a previous study.

Publication:

Hunnicutt, B. J. et al. (2016) A comprehensive excitatory input map of the striatum reveals novel functional organization. eLife. 5, e19103.

---

Data Sources:

Thalamic Projections : Hunnicutt, B. J. et al. (2014). A comprehensive thalamocortical projection map at the mesoscopic level. Nature Neuroscience. 17, 1276–1285.

Cortical Projections : Oh, S. W. et al. (2014). A mesoscale connectome of the mouse brain. Nature 508, 207–214.

---

Corticostriatal Data & Integrating with Other Network Data:

• Python/ & Matlab/

File Name	Folder	Purpose
1. jh_GetDensityDataFromWeb.py	Python	Access density data from AIBS API
2. jh_export2matlab4.py	Python	Get voxelized AIBS data out of python
3. jh_pImport2matlab2.m	Matlab	Get AIBS data aligned & prepped for analysis
4. jh_AllenInstituteBundleSubtraction.m	Matlab	Remove bundled projections from AIBS data
5. jh_consolidatingAIBSdatasets.m	Matlab	Group injection data by cortical origin
6. jh_corticostriatalFigures.m	Matlab	Generate figures for corticostriatal data alone
7. jh_voxelClustering_striatum.m	Matlab	Create striatal subdivisions based on convergent cortical inputs
8. jh_consolidatingThalamusData.m	Matlab	Get thalamic injections, group them, calculate coverage & nuclear coverage
9. jh_consolidatingAIBS_forNetworkAnalysis.m	Matlab	Generate data for network analyses
10. jh_assortedStriatumFigures.m	Matlab	Generate several example figures for methods and background

---

Thalamostriatal Data Generation:

• /Matlab/thalamostriatal/

File Name	Purpose
1. jh_segmentstriatum.m	Create manual striatum masks
2. jh_strRot.m	Manually select striatal landmarks used for alignment
3. jh_checkingStrPts.m	Check manually selected points
4. jh_createStrMaskedTiffs.m	Generate tiffs cropped by the striatum mask
--> WEKA Image Segmentation machine learning algorithm implemented via ImageJ	Select and train image subset, then apply WEKA machine learning algorithm to all images. Output => WEKA Probability Images for diffuse projection localization
5. jh_threshold_WEKA.m	GUI to manually select probability thresholds ( Requires: jh_threshold_WEKA.fig)
6. jh_WEKAprobToMask.m	Apply the selected thresholds to the probability masks
7. jh_finalProjMaskAdjustments_green.m	Manual correction of small errors in automated WEKA ML output for green channel (Requires: jh_finalProjMaskAdjustments_green.fig)
8. jh_finalProjMaskAdjustments_red.m	Manual correction of small errors in automated WEKA ML output for red channel (Requires: jh_finalProjMaskAdjustments_red.fig)
9. jh_createFinalProjMasks.m	Generate final projection masks that include manual adjustments and holes caused be traveling axons filled
10. jh_createFinalProjMasks_fixaddMaskMistake.m	Ran after jh_createFinalProjMasks.m to fix a small error.

See: /Matlab/thalamostriatal/README.md for implementation details.

---

Other Information:

Required Matlab Functionality:

• Image Processing Toolbox
• Statistics and Machine Learning Toolbox
• Medical Image Processing Toolbox by Alberto Gomez

Helper functions required for the code above:

• See: /Matlab/auxillary_funcitonsAndScripts/...

Collection of template masks, data, and settings required above:

• See: /Matlab/masks/...

Striatum alignment for thalamostriatal data:

• See: /Matlab/striatum_alignment/...
• Currently in the .gitignore - Need to ask Haining about putting this code in here with attribution