23.05.2025 Nora Strom
These files provide LOO GWAS analyses from the tic-disorder GWAS of the following publication:
Strom, N. I., Halvorsen, M. W., Grove, J., Ásbjörnsdóttir, B., Luðvígsson, P., Thorarensen, Ó., de Schipper, E., Bäckmann, J., Andrén, P., Tian, C., PGC TS Working Group, 23andMe Research Team, Als, T. D., Nissen, J. B., Meier, S. M., Bybjerg-Grauholm, J., Hougaard, D. M., Werge, T., Børglum, A. D., Hinds, D. A., … Mattheisen, M. (2025). Genome-Wide Association Study Meta-Analysis of 9619 Cases With Tic Disorders. Biological psychiatry, 97(7), 743–752. https://doi.org/10.1016/j.biopsych.2024.07.025

If you have any questions, contact Nora Strom (nora.strom@hu-berlin.de) or Manuel Mattheisen (manuel.mattheisen@dal.ca)

md5checksums for the sumstats files:
afd49f780291ca2e078349cb63c5f0fc  daner_TSTIC_NORDiC_iPSYCH_PGC_wo23andMe_180324.gz

The full GWAS contains 9619 European ancestry tic-disorder cases and 981,048 European ancestry controls. The GWAS excluding 23andMe data contains 8048 OCD cases and 39,710 controls. 


A Manhattanplot can be found in manhattan.nog2.*.pdf and a QQ-plot in qq.*.pdf

The meta-analysis was done using metal within ricopili.
The file is in ricopili-daner-format (see here for details: https://docs.google.com/document/d/1TWIhr8-qpCXB13WCXcU1_HDio8lC_MeWoAg2jlggrtU/):


CHR: chromosome	
SNP: rs-numbers	
BP: base-pair position on genome-build GRCh37
A1: Effect allele (not necessarily the minor allele!)	
A2: Allele 2	
FRQ_A_Ncases: A1 allele frequency of affected individuals, the max number of cases is appended to the header *
FRQ_U_Ncontrols: A1 allele frequency of unaffected individuals, the max number of controls is appended to the header *	
INFO: Imputation quality metric**	
OR: Effect size estimate with respect to A1	
SE: Standard error of the effect size estimate (referring to log(OR))	
P: p-value	
ngt: number of datasets where the SNP was genotyped directly (i.e. not imputed)***	
Direction: String of characters +,-,? Each character represents the direction of the estimated effect (+ is one direction, - the other, and ? means that this variant was missing in the original study).	
HetISqt: The I² statsitic for estimating heterogeneity among study effects. 
HetDf: Degrees of freedom for the heterogeneity test	
HetPVa: P-value of the test of heteogeneity among estimated effects in the studies	
Nca: The number of cases used in the test. 	
Nco: The number of controls in the overall meta-analysis. 	
Neff_half: The (half) effective sample size. For single cohorts this is defined as 4*Nca*Nco/(2*(Nca+Nco)). In meta analyses these numbers are summed over cohorts. 

* Note that not all individual studies provided separate allele frequencies for cases and controls.
** Also, not all of the individual datasets provided INFO scores. If they did not, the SNP infoscore was set to 1 for that study.
*** Not all individual datasets provided this information. If they did not, ngt was set to 0 for that study.