Prediction of hydrogen and carbon chemical shifts from RNA using database mining and support vector regression

摘要: The Biological Magnetic Resonance Data Bank (BMRB) contains NMR chemical shift depositions for over 200 RNAs and RNA-containing complexes. We have analyzed the 1H 13C shifts reported non-exchangeable protons of 187 these RNAs. Software was developed that downloads BMRB datasets corresponding PDB structure files, then generates residue-specific attributes based on calculated secondary structure. Attributes represent properties present in each sequential stretch five adjacent residues include variables such as nucleotide type, base-pair presence tetraloop types. central are used input to train a predictive model using support vector regression. These models can be predict new sequences. software tools, available stand-alone scripts or integrated into visualization analysis program NMRViewJ, should facilitate assignment and/or validation RNA shifts. In addition, our findings enabled re-calibration ring-current published high-resolution X-ray structural data guides.