Automated Assignment of Backbone NMR Data using Artificial Intelligence

Nuclear magnetic resonance (NMR) spectroscopy is a powerful method for the investigation of three-dimensional structures of biological molecules such as proteins. Determining a protein structure is essential for understanding its function and alterations in function which lead to disease. One of the major challenges of the post-genomic era is to obtain structural and functional information on the many unknown proteins encoded by thousands of newly identified genes. The goal of this research is to design an algorithm capable of automating the analysis of backbone protein NMR data by implementing AI strategies such as greedy and A* search. ar X iv :1 50 6. 05 84 6v 1 [ cs .A I] 1 8 Ju n 20 15 1 Nuclear Magnetic Resonance (NMR) Nuclear magnetic resonance is a phenomenon in which atomic nuclei absorb electromagnetic radiation at frequencies related to their chemical properties and the local molecular environment. Biophysicists use this property to gain structural knowledge of biomolecules, including proteins, DNA and RNA. NMR spectroscopy is currently the only method that allows the determination of atomic-level structures of large biomolecules in aqueous solutions similar to their in vivo physiological environments. Several types of NMR experiments can be used in the analysis of protein structures. In particular, essential information is provided by the chemical shifts of NMR-active nuclei present in proteins, including hydrogen and isotopes of carbon and nitrogen. The chemical shift is a quantifier for the deviation in the resonant frequency of a nucleus from its value in a structure-free environment, and therefore provides information on the local conformation. Determining the chemical shifts of all or most of the nuclei in a biomolecule is the first step in determining its structure. 1.1 NMR Assignment Methodology An important set of chemical shifts in a protein are those corresponding to the backbone nuclei, including the nitrogen, attached hydrogen, and the alpha and beta carbon atoms (Cα and Cβ) of each of the residues. The backbone residues constitute the building blocks of a protein chain (Figure 1). These checmical shift signals are measured using various NMR experiments, and then matched to the individual residues in the protein in a process called sequential assignment. Figure 1: HNCACB NMR experiment. A prerequisite of the assignment process is data collection using experiments that can provide information about the connectivities between neighboring residues [1]. One such experiment, called HNCACB, yields signals corresponding to the Cα and Cβ nuclei of one residue in the protein (residue i) , plus the Cα and Cβ signals of the immediately preceding residue (residue i− 1) (Figure 1). A second experiment, CBCA(CO) NH, can be used to yield the chemical shifts of the preceding residue only. These experiment are not independent, so they allow scientists to distinguish unambiguously between signals from residue i and residue i− 1. Analysis of all inter-residue connectivities allows the linking of signals from each backbone atom with signals from their preceding neighbor, creating a pattern of sequentially linked chemical shift values which reflects the sequential linear arrangement of the