A systematic two-sample and bidirectional MR process highlights a unidirectional genetic causal effect of allergic diseases on COVID-19 infection/severity

Table 3 Overall MR analyses with strong genetic instruments (p < 5E−08) for causal effects of peripheral blood hematological traits and count data of immune-related cells on COVID-19 infection/severity

Exposure	Outcome	No. of clumped SNPs^a	No. of SNPs in MRA^b	MR p-value^c	MR method^d	Heterogeneity	Pleiotropy	Directionality
Immune-related cells
CD4⁺ T/CD8⁺T	A2	21	19	7.21E−09	IVW (random effects)	Yes	No	True
	B2		20	5.36E−07	IVW (random effects)	Yes	No	True
	C2		20	4.63E−03	IVW (random effects)	Yes	No	True
CD4⁺ T	A2	1	1	3.17E−04	Wald ratio	–	–	True
	B2		1	3.42E−03	Wald ratio	–	–	True
	C2		1	3.38E−03	Wald ratio	–	–	True
CD56⁺ NK	A2	1	1	1.66E−01	Wald ratio	–	–	True
	B2		1	2.83E−03	Wald ratio	–	–	True
	C2		1	3.85E−01	Wald ratio	–	–	True
CD8⁺ T	A2	5	4	3.83E−03	IVW (random effects)	Yes	No	True
	B2		5	2.17E−02	IVW (random effects)	Yes	No	True
	C2		5	8.88E−02	IVW (random effects)	Yes	No	True
Hematological traits
HT	A2		1	4.92E−01	Wald ratio	–	–	True
	B2	1	1	2.14E−01	Wald ratio	–	–	True
	C2		1	4.42E−01	Wald ratio	–	–	True
LYMPH	A2	1	1	7.81E−01	Wald ratio	–	–	True
	B2		1	6.96E−01	Wald ratio	–	–	True
	C2		1	6.18E−01	Wald ratio	–	–	True
MCH	A2	4	3	4.86E−01	IVW (fixed effects)	No	No	True
	B2		3	4.40E−01	IVW (fixed effects)	No	No	True
	C2		3	2.50E−01	IVW (fixed effects)	No	No	True
MCV	A2	5	5	3.58E−01	IVW (fixed effects)	No	No	True
	B2		5	4.92E−01	IVW (fixed effects)	No	No	True
	C2		5	4.37E−01	IVW (fixed effects)	No	No	True
MONO	A2	3	3	9.39E−03	IVW (fixed effects)	No	No	True
	B2		3	5.11E−02	IVW (fixed effects)	No	No	True
	C2		3	3.63E−03	IVW (fixed effects)	No	No	True
PLT	A2	1	1	7.71E−02	Wald ratio	–	–	True
	B2		1	7.89E−01	Wald ratio	–	–	True
	C2		1	8.95E−01	Wald ratio	–	–	True
RBC	A2	1	1	1.78E−01	Wald ratio	–	–	True
	B2		1	7.21E−01	Wald ratio	–	–	True
	C2		1	8.12E−01	Wald ratio	–	–	True

–: The corresponding analysis is not available
^aNo. of clumped SNPs: number of independent genetic SNPs with a p-value < 5 × 10^–8 after clumping
^bNo. of SNPs in MRA: number of independent genetic SNPs used in the MR analysis for each pair of exposure and outcome
^cMR p-value: p-value of the most suitable MR method
^dThe most suitable MR analysis used in MR analysis. Bold values indicate that the MR p-value is less than 0.05, while italic values indicate that the MR p-value is marginally significant

ISSN: 1479-5876