Background: Myofibrillar myopathy (MFM) of unknown aetiology has recently been identified in Warmblood (WB) horses. In humans, 16 genes have been implicated in various MFM-like disorders. Objectives: To identify variants in 16 MFM candidate genes and compare allele frequencies of all variants between MFM WB and non-MFM WB and coding variants with moderate or severe predicted effects in MFM WB with publicly available data of other breeds. To compare differential gene expression and muscle fibre contractile force between MFM and non-MFM WB. Study design: Case-control. Animals: 8 MFM WB, 8 non-MFM WB, 33 other WB, 32 Thoroughbreds, 80 Quarter Horses and 77 horses of other breeds in public databases. Methods: Variants were called within transcripts of 16 candidate genes using gluteal muscle mRNA sequences aligned to EquCab3.0 and allele frequencies compared by Fisher's exact test among MFM WB, non-MFM WB and public sequences across breeds. Candidate gene differential expression was determined between MFM and non-MFM WB by fitting a negative binomial generalised log-linear model per gene (false discovery rate <0.05). The maximal isometric force/cross-sectional area generated by isolated membrane-permeabilised muscle fibres was determined. Results: None of the 426 variants identified in 16 candidate genes were associated with MFM including 26 missense variants. Breed-specific differences existed in allele frequencies. Candidate gene differential expression and muscle fibre-specific force did not differ between MFM WB (143.1 ± 34.7 kPa) and non-MFM WB (140.2 ± 43.7 kPa) (P =.8). Main limitations: RNA-seq–only assays transcripts expressed in skeletal muscle. Other possible candidate genes were not evaluated. Conclusions: Evidence for association of variants with a disease is essential because coding sequence variants are common in the equine genome. Variants identified in MFM candidate genes, including two coding variants offered as commercial MFM equine genetic tests, did not associate with the WB MFM phenotype.