Over the past decade, great progress in sequencing technologies and computational biology has revealed that the majority of the mammalian genome considered to be noncoding is rich in functional elements able to produce proteins. Many RNA molecules, mis-annotated as noncoding, actually harbor small open reading frames that are predicted to code for proteins. Some of those proteins have been verified to play critical roles in multiple biological processes. The lipid droplet (LD) is a unique cellular organelle, conserved from bacteria to humans, and is closely associated with cellular lipid metabolism and metabolic disorders. No noncoding RNA-coded proteins have been identified on LDs. Here, for the first time, we searched the organelle for their presence. After the enrichment of small proteins of LDs isolated from myoblasts, we used mass spectrometry coupled with our lab made protein database to identify LD-associated noncoding RNA-encoded proteins (LDANPs). A total of 15 new proteins were identified. One of them was studied further and termed LDANP1. LDANP1 was localized on LDs by imaging, cell fractionation, and immunogold labeling. Like LD resident proteins, LDANP1 was degraded by the proteasome. Using the CRISPR/Cas9-mediated genome editing technique, the endogenous expression of LDANP1 was validated. The stable expression of LDANP1 suppressed the accumulation of triacylglycerol in oleic acid treated myoblasts and inhibited the rescue of palmitate-inhibited insulin sensitivity by oleic acid. In summary, we report for the first time that translatable, nominally noncoding RNA-derived proteins, which are new and cannot be identified using current research methods, were associated with LDs and that among these, LDANP1 modulated lipid metabolism and insulin sensitivity. The discovery of noncoding RNA-encoded proteins on LDs paves a new way for the research of LDs and lipid metabolism.