Branching topology of the human embryo transcriptome revealed by entropy sort feature weighting

• DOI: 10.1101/2023.10.12.562031
• Publication Date: 2023-10-17T05:35:13Z
• Authors: Arthur Radley; Austin Smith

ABSTRACT Single cell transcriptomics (scRNA-seq) transforms our capacity to define states and reveal developmental trajectories. Resolution is challenged, however, by high dimensionality noisy data. Analysis therefore typically performed after sub-setting highly variable genes (HVGs). However, existing HVG selection techniques have been found poor agreement with one another, tend be biased towards expressed genes. Entropy sorting provides an alternative mathematical framework for feature subset selection. Here we implement continuous entropy sort weighting (cESFW). On synthetic datasets, cESFW outperforms in distinguishing state specific We apply six merged scRNA-seq datasets spanning human early embryo development. Without smoothing or augmenting the raw counts matrices, generates a high-resolution embedding displaying coherent progression from 8-cell post-implantation stages, delineating 15 distinct states. The highlights sequential lineage decisions during blastocyst development while unsupervised clustering identifies branch point populations. Cells previously claimed lack trajectory reside first branching region where morula differentiates into Inner Cell Mass (ICM) Trophectoderm (TE). quantify relatedness of pluripotent stem cultures types identify naïve primed marker conserved across culture conditions embryo. Finally, identifying specifically enriched dynamic expression formation, provide markers staging blastocyst. Together these analyses indicate that ability gene dynamics data can fail elucidate.