Antonio Lentini1, Huaitao Cheng2, JC Noble1, Natali Papanicolaou1, Christos Coucoravas1, Nathanael Andrews2, Qiaolin Deng3, Martin Enge2 and Björn Reinius1
1Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden. 2Department of Oncology and Pathology, Karolinska Institutet, Stockholm, Sweden. 2Department of Physiology and Pharmacology, Karolinska Institutet, Stockholm, Sweden.
This repository contains scripts and additional data needed to reproduce findings described in PAPER, PRE-PRINT.
Raw and processed data is available through ArrayExpress: [Smart-seq3], [Joint Smart-seq3+scATAC] and [Allelic dilution series].
x_escapees.tsv is compiled from (1, 2, 3, 4)
xy_homologs.tsv is modified from Table1 in (5)
X-chromosome inactivation and X-upregulation are the fundamental modes of chromosomewide gene regulation that collectively achieve dosage compensation in mammals, but the regulatory link between the two remains elusive and the X-upregulation dynamics are unknown. Here, we use allele-resolved single-cell RNA-seq combined with chromatin accessibility profiling and finely dissect their separate effects on RNA levels during mouse development. Surprisingly, we uncover that X-upregulation elastically tunes expression dosage in a sex- and lineage-specific manner, and moreover along varying degrees of X-inactivation progression. Male blastomeres achieve X-upregulation upon zygotic genome activation while females experience two distinct waves of upregulation, upon imprinted and random X-inactivation; and ablation of Xist impedes female X-upregulation. Female cells carrying two active X chromosomes lack upregulation, yet their collective RNA output exceeds that of a single hyperactive allele. Importantly, this conflicts the conventional dosage compensation model in which naïve female cells are initially subject to biallelic X-upregulation followed by X-inactivation of one allele to correct the X dosage. Together, our study provides key insights to the chain of events of dosage compensation, explaining how transcript copy numbers can remain remarkably stable across developmental windows wherein severe dose imbalance would otherwise be experienced by the cell.
R version 3.6.1 (2019-07-05)
| Package | Version | Source |
|---|---|---|
ArchR |
1.0.1 | Github (GreenleafLab/ArchR@968e442) |
Biobase |
2.44.0 | Bioconductor |
BiocGenerics |
0.30.0 | Bioconductor |
BiocParallel |
1.18.1 | Bioconductor |
biomaRt |
2.40.3 | Bioconductor |
boot |
1.3-23 | CRAN (R 3.6.1) |
circlize |
0.4.6 | CRAN (R 3.6.1) |
ComplexHeatmap |
2.0.0 | Bioconductor |
cowplot |
1.0.0 | CRAN (R 3.6.1) |
data.table |
1.12.2 | CRAN (R 3.6.1) |
DelayedArray |
0.10.0 | Bioconductor |
DESeq2 |
1.24.0 | Bioconductor |
GenomeInfoDb |
1.20.0 | Bioconductor |
GenomicRanges |
1.36.0 | Bioconductor |
ggbeeswarm |
0.6.0 | CRAN (R 3.6.1) |
ggplot2 |
3.2.1 | CRAN (R 3.6.1) |
hdf5r |
1.3.2.9000 | Github (hhoeflin/hdf5r@d38b053) |
IRanges |
2.18.1 | Bioconductor |
loomR |
0.2.1.9000 | Github (mojaveazure/loomR@1eca16a) |
magrittr |
1.5 | CRAN (R 3.6.1) |
MAST |
1.10.0 | Bioconductor |
matrixStats |
0.54.0 | CRAN (R 3.6.1) |
mclust |
5.4.5 | CRAN (R 3.6.1) |
princurve |
2.1.4 | CRAN (R 3.6.1) |
R6 |
2.4.0 | CRAN (R 3.6.1) |
RColorBrewer |
1.1-2 | CRAN (R 3.6.0) |
rtracklayer |
1.44.2 | Bioconductor |
S4Vectors |
0.22.0 | Bioconductor |
scater |
1.12.2 | Bioconductor |
scran |
1.12.1 | Bioconductor |
SingleCellExperiment |
1.6.0 | Bioconductor |
slingshot |
1.2.0 | Bioconductor |
SummarizedExperiment |
1.14.1 | Bioconductor |
tximport |
1.12.3 | Bioconductor |