Table 3. Comparison of significant metabolites in differential abundance GLM tests for APOE across Arivale and TwinsUK.
| Metabolite | Arivale E2 | Arivale E4 | TwinsUK E2 | TwinsUK E4 |
| 1-(1-enyl-stearoyl)-2-arachidonoyl-GPC (P-18:0/20:4) | 0.170 | –0.116 | ns | ns |
| 1-(1-enyl-stearoyl)-2-dihomo-linolenoyl-GPE (P-18:0/20:3)* | 0.190 | −0.126 | Not in Dataset | Not in Dataset |
| 1-(1-enyl-stearoyl)-2-oleoyl-GPE (P-18:0/18:1) | 0.163 | ns | 0.210 | ns |
| 1-docosapentaenoyl-GPC (22:5n3)* | ns | 0.108 | ns | 0.148 |
| 1-linoleoylglycerol (18:2) | 0.230 | 0.150 | ns | 0.120 |
| 1-oleoylglycerol (18:1) | 0.202 | 0.112 | ns | 0.150 |
| 1-stearoyl-2-arachidonoyl-GPE (18:0/20:4) | 0.166 | ns | 0.156 | ns |
| 1-stearoyl-2-linoleoyl-GPE (18:0/18:2)* | ns | 0.134 | ns | 0.128 |
| 1-stearoyl-GPE (18:0) | ns | 0.137 | ns | 0.160 |
| 1-stearyl-2-arachidonoyl-GPC (O-18:0/20:4)* | 0.232 | −0.110 | Not in Dataset | Not in Dataset |
| 2-hydroxybutyrate/2-hydroxyisobutyrate | ns | −0.118 | ns | −0.194 |
| 2-linoleoyl-GPE (18:2)* | ns | 0.110 | ns | 0.148 |
| 2-stearoyl-GPE (18:0)* | ns | 0.146 | ns | 0.134 |
| 3-(3-hydroxyphenyl)propionate | Not in Dataset | Not in Dataset | −0.166 | −0.126 |
| 4-methylcatechol sulfate | −0.197 | ns | 0.149 | ns |
| cholesterol | −0.172 | ns | −0.262 | 0.149 |
| diacylglycerol (16:1/18:2 [2], 16:0/18:3 [1])* | 0.209 | 0.112 | Not in Dataset | Not in Dataset |
| isoleucylvaline | Not in Dataset | Not in Dataset | 0.165 | 0.130 |
| linoleoyl-arachidonoyl-glycerol (18:2/20:4) [1]* | 0.312 | 0.150 | Not in Dataset | Not in Dataset |
| linoleoyl-linoleoyl-glycerol (18:2/18:2) [1]* | 0.221 | 0.177 | Not in Dataset | Not in Dataset |
| oleoyl-arachidonoyl-glycerol (18:1/20:4) [1]* | 0.304 | 0.148 | Not in Dataset | Not in Dataset |
| oleoyl-arachidonoyl-glycerol (18:1/20:4) [2]* | 0.315 | 0.133 | Not in Dataset | Not in Dataset |
| oleoyl-linoleoyl-glycerol (18:1/18:2) [1] | 0.208 | 0.156 | 0.166 | 0.132 |
| oleoyl-linoleoyl-glycerol (18:1/18:2) [2] | 0.206 | 0.127 | 0.158 | 0.133 |
| palmitoleoyl-arachidonoyl-glycerol (16:1/20:4) [2]* | 0.222 | 0.155 | Not in Dataset | Not in Dataset |
| palmitoleoyl-linoleoyl-glycerol (16:1/18:2) [1]* | 0.176 | 0.132 | Not in Dataset | Not in Dataset |
| palmitoyl dihydrosphingomyelin (d18:0/16:0)* | −0.224 | ns | −0.157 | ns |
| palmitoyl sphingomyelin (d18:1/16:0) | −0.220 | ns | −0.198 | 0.115 |
| palmitoyl-arachidonoyl-glycerol (16:0/20:4) [2]* | 0.309 | 0.101 | Not in Dataset | Not in Dataset |
| palmitoyl-linoleoyl-glycerol (16:0/18:2) [1]* | 0.197 | 0.107 | 0.147 | 0.119 |
| palmitoyl-linoleoyl-glycerol (16:0/18:2) [2]* | 0.220 | ns | 0.182 | ns |
| pyrraline | Not in Dataset | Not in Dataset | 0.172 | 0.123 |
| X - 11491 | Not in Dataset | Not in Dataset | 0.156 | 0.122 |
| X - 11795 | ns | 0.125 | ns | −0.137 |
| X - 13431 | ns | ns | 0.265 | 0.191 |
| X - 24065 | Not in Dataset | Not in Dataset | −0.204 | 0.168 |
Metabolites with pre-adjusted p < 0.05 for at least two out of four APOE β-coefficients in metabolite abundance GLMs across tests for Arivale and TwinsUK are reported. β-coefficient estimates are reported for metabolites with pre-adjusted p < 0.05, with bolded β-coefficients denoting pFDR < 0.1. Nonsignificance is indicated with ‘ns’ while metabolites missing from one of the two datasets are denoted. Metabolites are sorted alphabetically. Metabolite names ending in “*” indicate compounds not confirmed based on a standard but having high confidence in its identity, while those ending in “**” indicate compounds for which a standard is not available, but for which there is reasonable confidence in its identity or the information provided.