Table 1 Candidate proteins showing supporting evidence of potential causality for BD_ALL and BD_I by MR analysis, Steiger filtering analysis and Bayesian colocalization analysis.

Disease	Tissue	Protein	MR analysis				Steiger filtering analysis^a		Bayesian colocalization analysis^b		Evidence of causal association
Disease	Tissue	Protein	Method	nsnp	OR (95% CI)	p value	p value	Causal direction	PPH4	Colocalization	Evidence of causal association
BD_ALL	CSF	CTSF	Wald ratio	1	0.12(0.06,0.24)	9.95E−09	1.83E−07	Correct	1	Yes	Yes
	CSF	AGRP	Wald ratio	1	3.41(2.01,5.79)	5.75E−06	1.97E−09	Correct	0.996	Yes	Yes
	CSF	FRZB	Wald ratio	1	2.80(1.77,4.43)	1.01E−05	1.93E−13	Correct	0.993	Yes	Yes
	CSF	LRP8	Wald ratio	1	0.45(0.31,0.65)	1.64E−05	1.36E−17	Correct	0.980	Yes	Yes
	CSF	IL36A	Wald ratio	1	1.70(1.33,2.17)	1.92E−05	2.62E−49	Correct	0.982	Yes	Yes
	plasma	FLRT3	Wald ratio	1	0.70(0.58,0.84)	8.31E−05	6.99E−08	Correct	0.919	Yes	Yes
	CSF	MMEL1	Wald ratio	1	1.51(1.21,1.88)	2.71E−04	3.69E−82	Correct	0.793	No	No
	plasma	LYZ	Wald ratio	1	0.45(0.30,0.70)	2.86E−04	7.96E−23	Correct	0.769	No	No
BD_I	CSF	CTSF	Wald ratio	1	0.10(0.04,0.24)	6.28E−07	1.40E−07	Correct	0.999	Yes	Yes
	CSF	AGRP	Wald ratio	1	4.90(2.54,9.47)	2.25E−06	1.90E−09	Correct	0.998	Yes	Yes
	CSF	LRP8	Wald ratio	1	0.41(0.26,0.65)	1.20E−04	1.02E−17	Correct	0.903	Yes	Yes
	plasma	FLRT3	Wald ratio	1	0.66(0.53,0.83)	2.58E−04	6.29E−08	Correct	0.827	Yes	Yes
	plasma	LYZ	Wald ratio	1	0.38(0.23,0.65)	3.35E−04	7.14E−23	Correct	0.779	No	No

MR Mendelian randomization, BD_ALL Bipolar disorder, BD_I Bipolar I disorder.
^aA p value of less than 0.05 from the Steiger filter analysis was considered to be in the correct direction of the causal relationship from exposure to outcome.
^bA PPH4 higher than 0.80 from Bayesian colocalization analysis was considered strong evidence of colocalization.

Quick links

Search