---

This data collection contains spatially resolved single-cell transcriptomics datasets acquired using MERFISH on the developing human heart (13 PCW heart and 15 PCW ventricles) collected by a collaboration of the Chi Lab and the Center for Epigenomics at the University of California, San Diego.

The heart sections were imaged with 238 genes using MERFISH with a 22-bit Hamming distance 4, Hamming weight 4, binary code. The 22 bits are imaged in 11 hybridization rounds with two-color imaging in each round. This human heart panel included 238 genes for MERFISH imaging and 20 genes for sequential, two-color FISH imaging following the MERFISH run.

Description of the Data and file structure

• The dataset contains MERFISH images from a total of 4 experiments collected from two donors (3 x 13 PCW and 1 x 15 PCW). Each experiment was named with a sample_id that is composed of the imaging experiment id and replicate id. For instance, in ‘R1_R77_4C4’, R1 is replicate 1, and R77_4C4 is the imaging experiment id. R77_4C4, R78_4C12, and R78_4C15 correspond to the 13 PCW heart, while R123_N3S5 corresponds to the 15 PCW ventricles.

This folder contains the following files:

1. {Replicate}_{sample_id}*assigned_barcodes.csv: Decoded RNA spot location (global_x,global_y) in the unit of microns and their gene identity for each experiment and the cell id that each molecule belongs to for each experiment.
2. {Replicate}{sample_id}single_cell_raw_counts.csv. Cell-by-gene matrix for each experiment.
3. {Replicate}*{sample_id}_cell_metadata.csv: Metadata information of each sample. Contains volume (size of cell) information for each cell_id, as well as fov, global_x, global_y, and status from the ‘assignedFor each cell id, it contains the field
4. overall_merfish.h5ad: Scanpy object of the three replicate MERFISH experiments of a 13 PCW heart
5. ventricle_merfish.h5ad: Scanpy object of a ventricle subset of the 13 PCW MERFISH experiments
6. ventricle_15pcw_merfish.h5ad: Scanpy object of a MERFISH experiment of 15 PCW ventricles
7. overall_heart.rds: Seurat object of scRNA-seq from 9 PCW, 11 PCW, 13 PCW, and 15 PCW hearts
8. cardiomyocyte_compartment.rds: Seurat object of a cardiomyocyte subset of the overall_heart.rds object
9. mesenchymal_compartment.rds: Seurat object of a mesenchymal subset of the overall_heart.rds object
10. endothelial_compartment.rds: Seurat object of a endothelial subset of the overall_heart.rds object
11. neuronal_compartment.rds: Seurat object of a neuronal subset of the overall_heart.rds object
12. blood_compartment.rds: Seurat object of a blood subset of the overall_heart.rds object

Columns in the assigned_barcodes.csv table:

Barcode_id
The codebook index of the barcode, i.e. the gene. The gene column is more useful.

Fov
Which field of view (fov) the barcode is in

X,Y,Z
The pixel coordinates of the barcode within the FOV image. X and y range from 0 to 2048, with each pixel being ~108nm (the FOV is 220 microns). The Z coordinate is from 0 to 19, with each unit being 600nm.

Gene
The name of the gene for this barcode

Error_type, Error_bit
Error_type=0 means no error was corrected, -1 is a 1->0 error corrected, +1 is a 0->1 error corrected. Error_bit is the bit the error occurred in.

Status
The status of the barcode. If you want all barcodes ignoring cell segmentation, filter out barcodes with status “edge” and keep all the rest. If you want the barcodes kept after applying the cell segmentation and filtering, only use the barcodes with status “good”.

Cell_id
The cell ID this barcode was assigned to, or 0 if no cell.

global_x, global_y
Barcode position converted to global coordinates, in micron units.

Columns in the cell_metadata.csv table:

Cell_id
The assigned ID for each identified cell.

Fov
Which field of view (fov) the cell is in

Volume
The calculated size of the cell.

global_x, global_y
Cell position converted to global coordinates, in micron units.

Status
The status of the cell. Cells are filtered on volume and whether it contained detected barcodes. If you want the cells kept after applying filtering, only use the cells with status “ok”.

Details for h5ad files:

overall_merfish.h5ad - AnnData object containing the following information for each cell:
‘sample_id’ (experimental ID), ‘batch’ (experimental ID, same as sample_id for this object), ‘n_counts’ (number of transcript counts), ‘leiden’ (assigned cluster number), ‘zone_cluster’ (assigned cellular community number), ‘communities’ (assigned cellular community name), ‘complexity’ (number of unique populations in 150 micron cell zone), ‘populations’ (assigned population name), ‘purity’ (proportion of majority cell population within 150 micron cell zone)

ventricle_merfish.h5ad - AnnData object containing the following information for each cell:
‘sample_id’ (experimental ID), ‘batch’ (experimental ID, same as sample_id for this object), ‘n_counts’ (number of transcript counts), ‘leiden’ (assigned cluster number), ‘zone_cluster’ (assigned cellular community number), ‘communities’ (assigned cellular community name), ‘complexity’ (number of unique subpopulations in 150 micron cell zone), ‘populations’ (assigned population name), ‘purity’ (proportion of majority cell subpopulation within 150 micron cell zone), ‘subpopulations’ (assigned ventricular subpopulation name), ‘depth’ (depth within ventricular wall)

ventricle_15pcw_merfish.h5ad - AnnData object containing the following information for each cell:
‘n_genes’ (number of genes), ‘n_genes_by_counts’ (number of genes), ‘log1p_n_genes_by_counts’ (number of genes log transformed), ‘total_counts’ (number of transcript counts), ‘log1p_total_counts’ (number of transcript counts log transformed), ‘subpopulations’ (assigned ventricular subpopulation name)

Sharing/access Information

Links to other publicly accessible locations of the data:
https://cells.ucsc.edu/?ds=hoc

Was data derived from another source? No