id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-103150-e9q8e62v	Mishra, Shreya	Improving gene-network inference with graph-wavelets and making insights about ageing associated regulatory changes in lungs	2020-11-04		.txt	text/plain	8216	457	53	Just like gene-expression profile, inferred gene network could also be used to find differences in two groups of cells(sample) [13] to reveal changes in the regulatory pattern caused due to disease, environmental exposure or ageing. In order to test the hypothesis that graph-based denoising could improve gene-network inference, we first evaluated the performance of our method on bulk expression data-set. Our approach of graph-wavelet based pre-processing of mESC scRNA-seq data-set improved the performance of gene-network inference methods by 8-10 percentage (Fig. 2B) . Similarly in comparison to graph-wavelet based denoising, the other 7 methods did not provided substantial improvement in AUC for overlap among gene-network inferred by two data-sets of mESC (Fig. 2C , supplementary Figure S1B ). However, graph wavelet-based filtering improved the overlap between networks inferred from different batches of scRNA-seq profile of mESC even if they were denoised separately (Fig. 2C , supplementary Figure S1B ).	./cache/cord-103150-e9q8e62v.txt	./txt/cord-103150-e9q8e62v.txt