Genomic and molecular biology projects are generating knowledge data whose volume and complexity are unparalleled in this research area. In addition, data and knoweledge sources produced and used by research groups have terminological differences (synonyms, aliases and formulae), syntactic differences (file structure, separators and spelling) and semantic differences (intra- and interdisciplinary homonyms). In this context, data management techniques play a fundamental role for biological applications development because it offers adequate abstractions to desing, implement, access and manage data, in order to generate knowledge. In this work, we study the representation problems presentd in traditional languages. Following, we raise the main requiremants for a new conceptual data model specially conceived for molecular biology. Finally, we propose a new conceptual data model with special types of constructor tryng to solve some of the representation problems discurssed before. In addition, we formalize our proposed model using first-order logic and we use this logical description to infer some properties that may help database designer during the elaboration of database schema.