Poster abstracts
Poster number 99 submitted by Konrad Thorner
ELeFHAnt: An automated tool for harmonization and annotation of single cell RNA-seq data
Konrad Thorner (Cincinnati Childrens Hospital Medical Center, Developmental Biology), Praneet Chaturvedi (Cincinnati Childrens Hospital Medical Center, Developmental Biology)
Abstract:
Single cell sequencing has become an important method for understanding biological systems at an increasingly granular level. For single cell RNA-seq data specifically, one of the primary questions is using the transcriptome to determine cell identity. It is common to visualize such data and find clusters of cells with similarity in gene expression, but assigning each cluster a cell type is a much more open-ended task. Taking advantage of publicly available, annotated datasets in combination with supervised learning is a powerful approach for addressing this question. Ensemble Learning for Harmonization and Annotation of Single Cells (ELeFHAnt) provides an easy-to-use R package for users to annotate clusters of single cells, harmonize labels across single cell datasets to generate a unified atlas, and infer relationships among cell types between two datasets. It provides users with the flexibility of choosing between random forest and SVM (Support Vector Machine) based classifiers or letting ELeFHAnt apply both in combination to make predictions. ELeFHAnt is available on GitHub: https://github.com/praneet1988/ELeFHAnt
Keywords: machine learning, annotation, single cell RNA-seq