Motivation: Identifying protein enzymatic or pharmacological activities are important areas of research in biology and chemistry. Biological and chemical databases are increasingly being populated with linkages between protein sequences and chemical structures.There is now sufficient information to apply machine-learning techniques to predict interactions between chemicals and proteins at a genome scale. Current machine-learning techniques use as input either protein sequences and structures or chemical information. We propose here a method to infer protein-chemical interactions using heterogeneous input consisting of both protein sequence and chemical information. Results: Our method relies on expressing proteins and chemicals with a common cheminformatics representation. We demonstrate our approach by predicting whether proteins can catalyze reactions not present in training sets. We also predict whether a given drug can bind a target, in the absence of prior binding information for that drug and target. Such predictions cannot be made with current machine-learning techniques requiring binding information for individual reactions or individual targets. Availability and Contact: For questions, paper reprints, please contact Jean-Loup Faulon at jfaulon@sandia.gov. Additional information on the signature molecular descriptor and codes can be downloaded at:

Current SDSL-EPR methods allow measurement of dipolar distances in the 8-70 A range; however, the use of extrinsic probes complicates the interpretation of these distances in modeling macromolecular structure and conformational changes. The data presented here show that interprobe distances correlate only weakly with Cbeta-Cbeta distances, especially for distances that are on the order of the spin label tether lengths. Explicitly incorporating the spin label into the modeling process increases the experiment/model correlation 4-fold and reduces the distance error from 6 A to 3 A.

Electron paramagnetic resonance (EPR) is often used in the study of the orientation and dynamics of proteins. However, there are two major obstacles in the interpretation of EPR signals: (a) most spin labels are not fully immobilized by the protein, hence it is difficult to distinguish the mobility of the label with respect to the protein from the reorientation of the protein itself; (b) even in cases where the label is fully immobilized its orientation with respect to the protein is not known, which prevents interpretation of probe reorientation in terms of protein reorientation. We have developed a computational strategy for determining whether or not a spin label is immobilized and, if immobilized, predicting its conformation within the protein. The method uses a Monte Carlo minimization algorithm to search the conformational space of labels within known atomic level structures of proteins. To validate the method a series of spin labels of varying size and geometry were docked to sites on the myosin head catalytic and regulatory domains. The predicted immobilization and conformation compared well with