Table 2.

Sensitivity analysis of the causal association between PDEs proteins and psychiatric disorders

Exposure: outcome	MR-Eegger_Intercept	Egger_intercept_pvala	IVW_Cochrane_Q	IVW_Cochrane_Q_pvalb	MR-PRESSO Pvaluec	Steiger_testd
PDE1A: ASD	− 0.0069	0.6478	7.6507	0.8656	0.8810	TRUE
PDE2A: SCZ	0.0031	0.7621	23.5793	0.4274	0.4560	TRUE
PDE2A: TS	0.0190	0.4677	15.6579	0.8695	0.8547	TRUE
PDE3A: MDD	0.0008	0.8631	33.4772	0.0412	0.0533	TRUE
PDE4D: MDD	0.0027	0.5590	13.0823	0.4415	0.4583	TRUE
PDE4D: SCZ	− 0.0148	0.2230	18.0200	0.3227	0.3147	TRUE
PDE5A: TS	− 0.1120	0.0648	10.8504	0.1453	0.1750	TRUE
PDE7A: ADHD	− 0.0039	0.7274	16.6441	0.4787	0.4970	TRUE
OCD: PDE2A	0.0205	0.4210	6.1782	0.9861	0.9847	TRUE
OCD: PDE4D	0.0208	0.3611	18.0066	0.4552	0.5070	TRUE
OCD: PDE6D	− 0.0065	0.7844	6.8627	0.9613	0.9477	TRUE
AN: PDE9A	0.0077	0.6651	21.5561	0.8014	0.8070	TRUE
AD: PDE7A	− 0.0022	0.6965	67.5274	0.4247	0.4727	FALSE

Q Cochran’s Q statistics, IVW inverse-variance weighted, MR Mendelian randomization, MR-PRESSO MR pleiotropy residual sum and outlier, PDE phosphodiesterase, ASD autism spectrum disorder, SCZ schizophrenia, TS Tourette syndrome, MDD major depressive disorder, ADHD attention-deficit/hyperactivity disorder, OCD obsessive–compulsive disorder, AN anorexia nervosa, AD Alzheimer’s disease

^aThe MR–Egger intercept quantifies the effect of directional pleiotropy (P < 0.05, which means possible directional pleiotropy)

^bThe Cochrane—Q test quantifies the effect of heterogeneity (P < 0.05, which means possible heterogeneity, thus prioritizing “random—IVW” methods)

^cMR-PRESSO test quantifies the effect of horizontal pleiotropy (P < 0.05, which means possible horizontal pleiotropy)

^dMR-Steiger directionality test to assess the potential causal relationship (FALSE, which means an inverse causal link)