This repository provides data and relevant codes & toolbox used in our paper 'Development of Segregation and Integration of Functional Connectomes during the First 1000 Days', Cell Reports 43.5 (2024).
Content includes standalone software, source code, and demo data. Due to the large size of the analyzed data, we only provide a small portion of the data needed for validating the code. The project is structured into four parts corresponding to the major analyses in the article, including functional connectivity strength analysis, clustering coefficient & efficiency analysis, module analysis and gene expression analysis. Due to the file sizes exceeding the limit allowed by GitHub, the script relavent files and data can be found in https://pan.bnu.edu.cn/l/N1uFm7. The demo file size is 2.73G.
The dHCP data used in this study can be downloaded from https://biomedia.github.io/dHCP-release-notes/download.html and the BCP data can be downloaded from https://nda.nih.gov/edit_collection.html?id=2848 with requesting access.
Please use the “add path” method in MATLAB, and "install.packages()" method in R to add toolboxes and packages in the folder. These procedures are not time-consuming.
- smri preprocessing sMRI were preprocessed with following steps: average two images, reorient to fsl space, bias correction using N3, extract brain using skullStripping toolkit (v2) by Shifeng, and registration to UNC_4D_Volume_template. Details can be found in the manuscript.
- fmri preprocessing fMRI were preprocessed with following steps: drop first ten TRs and reorient, motion correction with SBRef, distortion correction, SBRef to anatomical registration, registration to BCP age-specific template and then to common template, smoothing, detrend & nuisance regressing, bandpass filtering, and scrubbing. Details can be found in the manuscript. The dependent tools are AFNI (Version AFNI_17.0.08), FSL (6.0.1) and SPM12. The workflow was meticulously crafted with the MICA pipeline as a guiding reference.
- FCS analysis FCS were computed at the voxel level. The gray matter mask were predefined by applying a threshold to the gray matter probability template of 6 months. The FCS (also within each distance bin) computation for each individual and the hub identification were computed using codes here. In order to explore the distribution of hubs across different functional systems, we mapped the Yeo's atlas on the 6-month infant brain. We first nonlinearly registered the MNI152 template to the BCP 24-month template using ANTs SyN algorithm (v2.3.4). And then, we linearly registered it to the 6-month template. The combined deformation was applied to Yeo's 7 network atlas in MNI152 space. Yeo's 7 network on 6-month infant brain can be found here.
- graph theory measures computation The graph theoretical measures on voxel-wise brain networks were executed using the Parallel Graph-theoretical Analysis (PAGANI) toolbox (v1.5).
- graph theory measures analysis The principal development axis of graph theoretical measures were analyzed using codes here.
- growth trajectory fitting The developmental trajectories were fitted by generalized additive mixed model (GAMM) in R (version 4.2.3). The dependent package is mgcv (version 1.8-42). The Gaussian random field (GRF) correction for the voxel level analysis were using the toolbox SeeCAT (20230518_beta).
- gene analysis Gene analysis and figure codes are all from Zhilei Xu's work "Meta-connectomic analysis maps consistent, reproducible, and transcriptionally relevant functional connectome hubs in the human brain" pusblised in the Communications Biology. The code link is here & here.
- visualization Visualization on cortical surface in this work were using the BrainNet Viewer (v1.7) software. The cortical surface file is here.
please cite us as follows:Li, Qiongling, et al. "Development of segregation and integration of functional connectomes during the first 1,000 days." Cell Reports 43.5 (2024).
Any question, please contact [email protected]