id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
10_1101-2020_08_28_271981	He, Jiahua	Full-length <em>de novo</em> protein structure determination from cryo-EM maps using deep learning	2021	26	.pdf	application/pdf	10020	1169	72	Full-length de novo protein structure determination from cryo-EM maps using deep learning structure types were predicted by a second DenseNet. Finally, the protein sequence was aligned to the main-chain according to the predicted Cα probabilities, amino acid types, and secondary structure amino acid type, and secondary structure type for each main-chain point, the target protein sequence The second network (i.e. DenseNet B) is used to predict the amino acid type and secondary structure type of a main-chain local dense point (LDP). Figure 3 shows a comparison of the predicted Cα models for the protein chains of different lengths The authors acknowledge professor Daisuke Kihara and his students Genki Terashi and Sai Raghavendra Maddhuri Venkata Subramaniya from Purdue University for providing their datasets. A New Protocol for Atomic-Level Protein Structure Modeling and Refinement Using Low-to-Medium Resolution Cryo-EM Density Maps. Figure 8: Protein models reconstructed by DeepMM and Phenix for the Chain A of 6DW1	./cache/10_1101-2020_08_28_271981.pdf	./txt/10_1101-2020_08_28_271981.txt