id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
10_1101-2020_01_29_925354	Bartoszewicz, Jakub M.	Interpretable detection of novel human viruses from genome sequencing data	2021	14	.pdf	application/pdf	13578	1392	61	As the state-of-the art approach for the openview detection of pathogens is genome sequencing (5, 6), it learning (17) to predict host range for a small set of three wellstudied species directly from viral sequences. predicting whether a new virus can potentially infect humans. boundary separates human viruses from other DNA sequences generated the reads from the genomes of human-infecting constituting reads yields a prediction for the whole sequence. predictions from all the reads originating from a given genome In the Fig. 1 we present example filters, visualized as "maxcontrib" sequence logos based on mean partial Shapley values prediction directly from next-generation sequencing reads Three receptor-binding domains (RBDs) are colored in blue, white and red according to the predicted infectious potential of the corresponding genomic sequence. Interpretable detection of novel human viruses from genome sequencing data Interpretable detection of novel human viruses from genome sequencing data	./cache/10_1101-2020_01_29_925354.pdf	./txt/10_1101-2020_01_29_925354.txt