Home

Welcome to the PLSCaEF wiki!

Abstract:

Decoding EF components with biological data

Run the codes according to the order:

Step_1st_PrepareData

Step_1st_behavioral_data_combine.R

Get EF-related data, including 12 tasks and 23 variables

Step_2nd_get_fc_vector.m

Transfer FC matrixes to vectors

Step_3th_fd_sex_age_set_calculate_merge.m

Calculate mean FD across runs. code site (different scanners are treated as different batches), merge FD,sex,age and set

Step_4th_combat.py

Using combat to control the effect of sites

Step_5th_merge_ef_fc_cov.m

Merge EF, FC,d and covariables

Step_2nd_PLSca

Step_1st_PLSCa.py

PLSCa modeling, with 2 CV, 101 times repetition, and 1000 times permutation

Step_2nd_PLSca_Sig.m

Calculate the significant level of the correlation between each pair of components and calculate the covariance explained by each pair of components. According to the covariance explained, the first two pairs of components were considered in the following analyses.

Step_3rd_Features_sig.m

Calculate the significant level of each feature, including brain and behavioral sides. The signs of features' weights are re-assigned to keep them consistent across different repetitions.

Step_4th_predict_psy.m

Compare the differences of two EF component scores between the healthy group and 8 psychiatric diagnosed gourps.