Abstract
forthcoming.
Team
| Role | Name |
|---|---|
| Project Lead | Audrey C. Luo |
| Faculty Lead | Theodore D. Satterthwaite |
| Analytic Replicator | Steven L. Meisler |
| Collaborators | Valerie J. Sydnor, Aaron Alexander-Bloch, Joëlle Bagautdinova, Deanna M. Barch, Dani S. Bassett, Christos Davatzikos, Alexandre R. Franco, Jeff Goldsmith, Raquel E. Gur, Ruben C. Gur, Fengling Hu, Marc Jaskir, Gregory Kiar, Arielle S. Keller, Bart Larsen, Allyson P. Mackey, Michael P. Milham, David Roalf, Golia Shafiei, Russell T. Shinohara, Leah H. Somerville, Sarah M. Weinstein, Jason Yeatman, Matthew Cieslak, Ariel Rokem |
Project Timeline
| Project Start Date | February 2024 |
|---|---|
| Current Project Status | In prep |
Code and Communication
| Github Repository | https://github.com/PennLINC/luo_wm_dev/tree/main |
|---|---|
| Slack Channel | #luo_wm_dev |
Datasets
- RBC PNC (health exclude) as discovery datasets, HCP-D (replication), and HBN (replication)
Conference Presentations
- Poster presented at The Organization for Human Brain Mapping Annual Meeting, June 2024
- Poster presented at Flux Congress, September 2024
- Poster to be presented at The Society of Biological Psychiatry Annual Meeting, April 2025 as a Predoctoral Travel Awardee
- Talk entitled Axes of Hierarchical Brain Development to be given at the Gradients of Brain Organization Workshop, Brisbane, Australia, June 2025
The project directory on CUBIC is /cbica/projects/luo_wm_dev .
- Code for the final manuscript is in
/cbica/projects/luo_wm_dev/two_axes_manuscript - Code for the analytic replication is in
/cbica/projects/luo_wm_dev/two_axes_replication. The replication directory is almost identical to the manuscript.
~/two_axes_*/code |
|
|---|---|
~/two_axes_*/input |
|
~/two_axes_*/output |
|
~/two_axes_*/software |
|
~/atlases |
|
~/SAaxis |
|
~/input/ |
Overview of Analytic Workflow
| Step | Task | Notes |
|---|---|---|
| 1 | Get static data from PMACS | |
| 2 | Do initial sample selection | |
| 3 | Use babs to run qsirecon and pyAFQ to get tract profiles |
|
| 4 | Get tract profiles from babs |
datalad get the merged babs output and copy over the tract profiles csv to input/<dataset>/derivatives/tract_profiles_dti |
| 5 | Prepare data for final sample selection | prep_data: finds subjects that are missing data and saves out a list of them for exclusion. This step needs to be done before final sample selection |
| 6 | Do final sample selection | after prep_data , I did my final sample construction and excluded subjects that failed babs, etc. |
| 7 | Harmonize multi-site data using CovBat | Only HCP-D and HBN need to be harmonized within each dataset |
| 8 | Fit GAMs | Compute magnitude of the age effect and age of maturation, as well as GAM fits for each age |
| 9 | Tract-to-cortex mapping | Lots of steps here! Goal is to get tract-to-cortex probability maps for each dataset and parcellate to HCP-MMP atlas |
| 10 | Significance testing with NEST: deep-to-superficial | Are age effects enriched in superficial compared to deep tract regions? (For Figures 1 and 2) |
| 11 | Significance testing with NEST: tract-to-cortex | Age age effects different on each end of callosum motor and IFOF? (For Figures 3 and 4) |
| 12 | Significance testing with spun t-test: delta-delta | Are the S-A ranks different between tracts with large vs. small differences in age of maturation? (For Figure 5) |
| 13 | Significance testing with spin test: age of maturation vs. S-A rank | Collapsing across tracts, does the age of maturation of superficial tract regions correlate with S-A axis rank? (For Figure 6) |
| 14 | ||