Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2023 Apr 3;42(1):99–108. doi: 10.1038/s41587-023-01728-5

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2023

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Extended Data Fig. 10 — (a, b) Scatter plots showing the total time (a) and the training speed (b) of cellDancer when applying multiprocessing. We tested the parallel speedup ratio of cellDancer by increasing job numbers from 1 to 30. We applied full analysis of cellDancer to 2,159 genes in 18,140 cells (the hippocampal dentate gyrus neurogenesis dataset) with the default parameters and calculated the runtime and speed of different job numbers. The evaluation of all the algorithms and the speedup ratio analysis was performed on a 2.7 GHz 24-Core Intel Xeon W processor. Total runtime decreases from 286 to 36 minutes when adding job numbers from 1 to 30 and reaches saturation at 15 jobs with 40 minutes. cellDancer has a feasible runtime of 53 genes per minute using 15 jobs. The training speed (number of genes per unit time) increases with the number of jobs. (c) Bar plot showing the total time of the comparison between velocyto, scVelo, DeepVelo, VeloVAE, and cellDancer. We compared the runtime of cellDancer with velocyto, scVelo, DeepVelo, and VeloVAE. The benchmark is based on 18,140 cells and 2,159 genes in the hippocampal dentate gyrus neurogenesis dataset with the default parameters. We set the number of jobs (threads) to 15 for scVelo, DeepVelo, VeloVAE, and cellDancer. cellDancer shows a comparable running time with the other two deep learning algorithms.