A methodology for structurally representing the molecules of three Colombian vacuum residues (538+ degrees C) and one Mexican atmospheric residue (300+ degrees C) is reported. Information obtained by Fourier transform ion cyclotron resonance mass spectrometry coupled to positive electrospray ionization, negative electrospray ionization, and positive atmospheric pressure photoionization sources and conventional standardized analytical methods was used for molecular representation of the samples. The generation of molecules was performed by a Monte Carlo technique, obtaining a set of representative structures for the global representation through attributes for each residue. The structural attributes considered are CH, CH2, and CH3 in paraffinic chains, number of naphthenic rings, CH2 and CH naphthenic number of aromatic rings, aromatic carbon type, cata- and peri-condensed carbons, number of sheets in asphaltenes, -SH, aromatic S (thiophene), aromatic N (pyridine and pyrrole), and -NH2. Each attribute in the residues can be represented by a probability density function (PDF), which is optimized for the purpose of adjusting the structures of residues and their composition to the experimental data. The PDFs for aromatics and nitrogen and sulfur components were obtained by ultrahigh mass resolution data. As a result, 150 molecules per each residue were obtained, and the mode of representation was single-feed (only one feed is characterized at the same time). The bulk properties for each residue were in good agreement with the experimental structural information.