Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: ccImpute: an accurate and scalable consensus clustering based algorithm to impute dropout events in the single-cell RNA-seq data

Fig. 3

The Adjusted Rand Index (ARI) scores of K-means clustering on t-SNE reduced scRNA datasets vs. imputation method. Clustering performance is a strong indicative of improved downstream performance, as long as the data is not heavily biased as a result of imputation. t-SNE is a non-linear technique, and this metric aims to measure impact of the imputation on correcting the non-linear patterns in the data. The range of possible values is in interval \([-1,1]\), with higher value indicating better performance. ccImpute is the best performing approach on all datasets, and the only algorithm that did not hurt the performance of t-SNE algorithm on any of the datasets. scImpute, DCA, and DeepImpute only work with raw unnormalized datasets and cannot impute the Usoskin dataset. Further, scImpute and DrImpute timed out on the larger datasets

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us