Reading Impairment in the Healthy Brain Network

In this project, we provide a summary description of participants with reading impairment in the healthy brain network dataset and we also do predictive modeling to assess which phenotypic data (including clinical questionnaires) are most successful in classifying participants with a reading impairment from other groups.

See here for a full assessment list administered as part of the Healthy Brain Network Initiative

Check out this presentation to see some of the preliminary results.

The accompanying notebook contains all of the figures, you can also run this notebook by following the steps outlined below.

The data are stored on google drive

Note: this is a work-in-progress, there are more models running (they sometimes take 1-2 days of computing time), I will update the notebook and the presentation as more results come in.

First Steps

• Clone Repo

Clone the repo to your local path

git clone [email protected]:maedbhk/hbn_reading.git`

• Activate Virtual Environment In this project, we're using conda for virtual environment and python package management

• To install and activate conda environment, run the following commands in the top-level directory of your cloned repo

# navigate to top-level directory of repo
$ cd ../hbn_reading

# create conda environment using .yml file
$ conda env create -f environment.yml 

# activate the conda virtual environment
$ conda activate hbn_reading # each virtual env has its own unique name (look inside the environment.yml file)

# install editable package (make sure virtual env is activated) so that you can access `src` packages
$ pip install -e .

• Activate jupyter notebook kernel

To run jupyter notebook using modules installed in virtual env, run the following command in top-level directory of repo

# make sure you are in top-level directory of repo
$ cd ../hbn_reading

# make sure conda env is activated
$ conda activate hbn_reading

# create kernel (virtual env for jupyter notebook)
$ ipython kernel install --name "reading" --user # you can assign any name

Notebooks

• Main results are stored here: notebooks/figures-Reading_Impairment.ipynb

To access this jupyter notebook, run the following commands

# first, make sure your virtual environment is activated
$ conda activate hbn_reading

# navigate to the notebooks directory
$ cd ../hbn_reading/notebooks

# open the jupyter notebook
$ jupyter notebook figures-Reading_Impairment.ipynb

# once you have opened the notebook, make sure you choose the correct kernel "reading" as it contains all of the python packages for this project

Project Organization

Data

Note: hbn_reading data folder is stored on Google Drive: I suggest downloading the google drive folder (it's not that big) and work locally OR there are python plugins that allow you to access google drive folders from jupyter notebooks

├── data
│   ├── Child-features-preprocessed.csv   <- All features (preprocessed) for Child questionnaires
│   ├── Child-features-raw.csv            <- All features (raw) for Child questionnaires
│   ├── Parent-features-preprocessed.csv  <- All features (preprocessed) for Parent questionnaires
│   ├── Parent-features-raw.csv           <- All features (raw) for Parent questionnaires
│   ├── Teacher-features-preprocessed.csv <- All features (preprocessed) for Teacher questionnaires
│   ├── Teacher-features-raw.csv          <- All features (raw) for Teacher questionnaires
│   ├── item-names-cleaned.csv            <- Dictionary linking questionnaire columns to question keys
│   ├── Data_Dictionaries.zip             <- Zipped folder containing data dictionaries (i.e. itemized breakdown of questions)
│   └── models                            <- Top-level folder containing model output

Code

├── README                                <- The top-level README for developers using this project
├── environment.yml                       <- Required python libraries
├── notebooks                             <- Notebooks folder
│   ├── figures-Reading_Impairment.ipynb  <- Example notebook
├── references                            <- Project-relevant materials
│   ├── Assessment_List_Jan2019.xlsx      <- List of all assessments administered by healthy brain network
├── src                                   <- src folder (customized python scripts)
│   ├── dataset_summary.py                
│   ├── model_summary.py         

---

Project based on the cookiecutter data science project template. #cookiecutterdatascience