Abstract
An increasing number of density maps of macromolecular structures, including proteins and DNA/RNA complexes, have been determined by cryo-electron microscopy (cryo-EM). Although lately maps at a near-atomic resolution are routinely reported, there are still substantial fractions of maps determined at intermediate or low resolutions, where extracting structure information is not trivial. Here, we report a new computational method, Emap2sec+, which identifies DNA or RNA as well as the secondary structures of proteins in cryo-EM maps of 5 to 10 Å resolution. Emap2sec+ employs the deep Residual convolutional neural network. Emap2sec+ assigns structural labels with associated probabilities at each voxel in a cryo-EM map, which will help structure modeling in an EM map. Emap2sec+ showed stable and high assignment accuracy for nucleotides in low resolution maps and improved performance for protein secondary structure assignments than its earlier version when tested on simulated and experimental maps.
Highlights
An increasing number of density maps of macromolecular structures, including proteins and DNA/RNA complexes, have been determined by cryo-electron microscopy
The number of determined cryoEM maps deposited in the public database, EMDB3, is growing exponentially; and the fraction of high resolution maps among them shows a steady increase
Recent methods benefit from the strong image recognition capabilities of deep learning to extract the local and global density features of EM maps[15,16]
Summary
An increasing number of density maps of macromolecular structures, including proteins and DNA/RNA complexes, have been determined by cryo-electron microscopy (cryo-EM). We report a new computational method, Emap2sec+, which identifies DNA or RNA as well as the secondary structures of proteins in cryo-EM maps of 5 to 10 Å resolution. In a map of intermediate resolution (~4–10 Å), protein secondary structure elements can be often detected even in cases where tracing a full sequence is difficult Tools for this task include those which detect typical local densities that correspond to an α-helix and βsheet[13,14]. We have extended the method to detect both protein secondary structure elements and nucleic acids in EM maps of intermediate resolution. Emap2sec+ showed high accuracy for nucleotide detection while maintaining comparable, if not achieving better, protein secondary structure detection performance to Emap2sec
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
