Unsupervised detection of biologically sound clusters in single-cell expression data

VERCESI, MARIA LUZ; BERARDINO, ARIEL; VEGA WAICHMAN, TOMAS; GIACOMINI, DAMIANA; RASETTO, NATALI BELEN; ARLOTTA, PAOLA; SCHINDER, ALEJANDRO; A. CHERNOMORETZ

Bethesda, Maryland

Simposio; FIC Global Brain Virtual Network Meeting; 2021

Fogarty International Center

Single-cell RNAseq assays provide a glimpseinto cellular transcriptional landscapes. Structures recognized inlow-dimensional manifolds probed by this technology allow to explorecell variability, identify new cell-types and uncover developmentalpathways at molecular levels. However, detecting patterns at theright scale to unveil relevant biology is not an easy task. Usually,manual curation is needed in order to identify the right resolutionat which cells should be grouped together. In this preliminary study,we propose an unsupervised method that capitalizes on biologicalinformation to uncover biologically meaningful cell clusters. Usingour methodology in a public and annotated dataset (Hochgerner2018) weshow that the proposed unsupervised pipeline produces robust resultsand can recapitulate the annotated cell type provided by the authorsof the original publication.p { margin-bottom: 0.1in; direction: ltr; line-height: 120%; text-align: left; orphans: 2; widows: 2; background: transparent }a:link { color: #0000ff; text-decoration: underline }