Abstract
METHODS AND RESULTS: Plasma fatty acid patterns were derived from 26 plasma fatty acids using k-means cluster analysis. Four clusters were identified, each with a distinct fatty acid profile. Cluster 1 included high proportions of linoleic acid (LA) and low proportions of stearic acid (SA); cluster 2 was higher in n-3 polyunsaturated fatty acids and SA; the profile of cluster 3 was higher in very-long-chain saturated fatty acid (VLCSFA) and lower in α-linolenic acid (ALA) (cluster 3); while cluster 4 was higher in fatty acids related to de novo lipogenesis and 20:3n-6 and lower in LA (cluster 4). In general, cluster 4 was associated with adverse metabolic profile and higher metabolic risk (p < 0.033). Clusters 2 and 3 were associated with healthier and protective phenotypes (p < 0.033).
CONCLUSION: Distinct fatty acid patterns were identified which were related to demographics, dietary habits, and metabolic profile. A pattern higher in VLCSFA and lower in ALA was associated with healthier metabolic outcome.
| Original language | English |
|---|---|
| Pages (from-to) | 2043-2052 |
| Number of pages | 10 |
| Journal | Molecular nutrition & food research |
| Volume | 60 |
| Issue number | 9 |
| DOIs | |
| Publication status | Published - 1 Sep 2016 |
Keywords
- k-means
- Metabolic risk
- Pattern analysis
- Plasma fatty acids
- Very-long-chain saturated fatty acid