Longitudinal multi-omics study reveals common etiology underlying association between plasma proteome and BMI trajectories in adolescent and young adult twins

Table 3 Mixed-effects model-derived estimates of proteins significantly associated with the changes (slope) of body mass index over 10 years of follow-up in FinnTwin12 participants

					P-value
Protein description	UniProt ID	Estimate	SE	t-value	Nominal	Bonferroni
Proteoglycan 4	Q92954	0.27	0.04	7.3	9.80E − 13	4.30E − 10
Complement factor H	P08603	0.25	0.04	6.6	7.70E − 11	3.40E − 08
Cholinesterase	P06276	0.24	0.04	5.6	3.00E − 08	1.30E − 05
Complement component C7	P10643	− 0.20	0.04	− 5.2	2.40E − 07	1.10E − 04
Sex hormone-binding globulin	P04278	− 0.24	0.05	− 5.1	4.50E − 07	2.00E − 04
C4b-binding protein alpha chain	P04003	0.18	0.04	4.8	1.70E − 06	7.60E − 04
Serum amyloid P-component	P02743	0.18	0.04	4.6	6.40E − 06	2.80E − 03
Complement factor I	P05156	0.16	0.04	4.5	1.00E − 05	4.40E − 03
Complement factor B	P00751	0.16	0.04	4.3	1.90E − 05	8.10E − 03
Phosphatidylcholine-sterol acyltransferase	P04180	0.16	0.04	4.3	2.30E − 05	9.90E − 03
Secreted phosphoprotein 24	Q13103	− 0.17	0.04	− 4.3	2.20E − 05	9.90E − 03
C-reactive protein	P02741	0.16	0.04	4.2	2.70E − 05	1.20E − 02
Properdin	P27918	0.15	0.04	4.0	7.10E − 05	3.10E − 02
Inter-alpha-trypsin inhibitor heavy chain H2	P19823	− 0.15	0.04	− 4.0	8.10E − 05	3.60E − 02

Protein estimates whose abundance were significantly associated with changes in body mass index (i.e., slope of body mass index) are shown if Bonferroni-corrected coefficient null-test p-values were less than 0.05. Mixed-effects models included as covariates: the intercept of body mass index, age at blood sampling, age squared, sex, and interactions of sex with age and age squared. Random effects included the zygosity indicator variable and family identifiers. The slope of body mass index was z-scored; fixed effects estimates indicate effect on 1 sd change
se Standard error

ISSN: 1741-7015