Abstract
[Display omitted]
•Average lamb overall liking of Chinese (CN) and New Zealand (NZ) consumers differed.•Similar clusters were generated for CN and NZ consumers using overall liking scores.•Clusters’ liking scores were associated with metabolomic and lipidomic compounds.
Relationships between overall liking scores for cooked lamb from Chinese (n = 158) and New Zealand (n = 156) consumers, and metabolite and lipid profiles were evaluated. Consumers assessed meat from 6 to 8-month-old lambs of composite genetics fed chicory (CHIC) or grass (GRASS), and from 12 month–old Merino lambs fed a mixed pasture (PMER). On average, Chinese consumers rated the overall liking of all types of lamb similarly, while New Zealand consumers preferred meat from CHIC over PMER. However, three clusters with similar preferences were obtained for both Chinese and New Zealand consumers based on their overall liking scores. In Cluster-1 with a preference for GRASS, overall liking for Chinese and NZ consumers was positively associated with umami compounds, ortho- and pyrophosphates (related to water holding capacity of meat), triglycerides (TG) with<50 carbons (C50) and phospholipids with polyunsaturated fatty acids (PUFA); but negatively associated with amino acids and TG with > C50 with saturated (SFA) and monounsaturated (MUFA) fatty acids. In Cluster–2 with a preference for CHIC, overall liking for both types of consumers was positively associated with TG with > C50 with PUFA, and phospholipids with PUFA, but negatively associated with umami compounds, ortho- and pyrophosphates and L-anserine. In Cluster–3 with a preference for PMER, overall liking for Chinese and NZ consumers was positively associated with amino acids, ortho- and pyrophosphates, L-anserine, umami compounds, TG with > C50 with SFA and MUFA and phospholipids that contain C16:0, C16:1, C18:0 and C18:1; but negatively associated with phospholipids with PUFA and TG with < C50 that contain PUFA. Overall, the liking of lamb meat between Chinese and New Zealand consumers differed, but similar clusters were generated based on their overall liking scores. The clusters were characterized by different associations of the consumer overall liking scores with cooked meat metabolome and lipidome profiles.