Abstract
BackgroundSingle cell transcriptomics is critical for understanding cellular heterogeneity and identification of novel cell types. Leveraging the recent advances in single cell RNA sequencing (scRNA-Seq) technology requires novel unsupervised clustering algorithms that are robust to high levels of technical and biological noise and scale to datasets of millions of cells.ResultsWe present novel computational approaches for clustering scRNA-seq data based on the Term Frequency - Inverse Document Frequency (TF-IDF) transformation that has been successfully used in the field of text analysis.ConclusionsEmpirical experimental results show that TF-IDF methods consistently outperform commonly used scRNA-Seq clustering approaches.
Highlights
Single cell transcriptomics is critical for understanding cellular heterogeneity and identification of novel cell types
Each of the 36 clustering algorithms described in the Methods section was run on 2-class synthetic mixtures of 1,000 cells sampled in different ratios from six pairs of immune cell types as described in Experimental setup
Each plot shows the median of the corresponding measure as the middle horizontal line, along with mean values as the middle points connected by lines across methods
Summary
Single cell transcriptomics is critical for understanding cellular heterogeneity and identification of novel cell types. Leveraging the recent advances in single cell RNA sequencing (scRNA-Seq) technology requires novel unsupervised clustering algorithms that are robust to high levels of technical and biological noise and scale to datasets of millions of cells. The recent advances in single cell RNA sequencing (scRNA-Seq) technologies promise to unveil novel cell types and uncover subtle regulatory processes that are undetectable by analyzing bulk samples. Droplet-based technologies such as the Chromium Megacell commercialized by 10x Genomics can quickly and inexpensively generate scRNA-Seq expression profiles for up to millions of cells. The large amounts of data and high levels of noise render many unsupervised clustering methods developed for bulk gene expression data [1] unusable, prompting the development of a new generation of clustering tools.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
