Figure 3
Gene Ranking and results of scRNA-seq data. Bubble color in figure b and line color in figure c indicates the method used, which corresponds to the color of x-axis labels in figure d. Part of abbreviation used the figure: Brennecke’s highly variable method (HV), M3Drop (M3), depth-adjusted negative binomial model (NB), chi-square statistic (CHI2), F-statistic (F). (A) Modified Pipeline of MATTE used in gene ranking: each phenotype pairs will be evaluated separately and SNR of each module will be taken as a gene score. Results from each turn will be added as final score. (B) Comprehensive performance of methods in this task. (C) With the number of marker genes increases, the performance of unsupervised learning task changes. (D) Average ratio of prediction of methods.

Gene Ranking and results of scRNA-seq data. Bubble color in figure b and line color in figure c indicates the method used, which corresponds to the color of x-axis labels in figure d. Part of abbreviation used the figure: Brennecke’s highly variable method (HV), M3Drop (M3), depth-adjusted negative binomial model (NB), chi-square statistic (CHI2), F-statistic (F). (A) Modified Pipeline of MATTE used in gene ranking: each phenotype pairs will be evaluated separately and SNR of each module will be taken as a gene score. Results from each turn will be added as final score. (B) Comprehensive performance of methods in this task. (C) With the number of marker genes increases, the performance of unsupervised learning task changes. (D) Average ratio of prediction of methods.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close