Key APA output file _ readme

Pyxa Data File ReadMe

Description:

After the initial decoding steps, APA (Asynchronous Primary Analysis) is initiated to carry out steps such as cell segmentation, transcript assignment, UMAP clustering etc. After a successful APA run, Cell_by_gene.csv, Cell_metadata.csv, Cell_assigned_gene.csv, pyxa_studio.csv and segmentation_geometries.parquet files are generated. These files can be imported into the Pyxa instrument for visualization or downstream data analysis.

Cell_by_gene.csv

Column descriptions:

CellID: Unique identifier for a single cell.

Gene expression target (AATK, ABL1, …, etc): The number of transcripts observed for a given gene target for a given cell. All transcripts not assigned to a cell are show with a “cell_ID” value of 0.

Negative probes (i.e. ERCC): Negative probes, which do not match any sequence within the transcriptome or genome, are included. These can be used to assess background levels.

System Control codes (i.e. Blank##): Blank Control codes and codes which do not have any physical probe associated with them. This can be used to monitor decoding accuracy.

Cell_metadata.csv

Column descriptions:

FOV: The Field Of View (FOV) the cell is in. All FOVs from the sample are contained within this Cell Metadata file.

Volume_um3: Cell volume based on cell sizes measured in um3.

Cell_id: Unique identifier for a single cell.

ROI: Region of Interest based on annotation during experimental set up.

X_pixels, Y_pixels, and Z_pixels: The x,y and z positions of centroid of a cell within the FOV, measured in pixels. The pixel edge length at x and y is 229.997nm. Thus, to convert to microns multiply the pixel value by 0.229997 um per pixel for x and y dimension. For z dimension, the pixel edge length at is 500 nm. Thus, to convert to microns multiply the pixel value by 0.5 um per pixel for z dimension.

X_um, Y_um and Z_um: see description for X_pixels, Y_pixels, and Z_pixels.

Cell_assigned_gene.csv

Column descriptions:

FOV: The Field Of View (FOV) the transcript is in. All FOVs from the sample are contained within this Cell Metadata file.

Gene: Gene that is detected in a Pyxa run.

X_pixels, Y_pixels, and Z_pixels: The x,y and z coordinate (in pixels) of this transcript within the FOV, measured in pixels. The pixel edge length is 229.997nm. Thus, to convert to microns multiply the pixel value by 0.22997 um per pixel. For z dimension, the pixel edge length at is 500 nm. Thus, to convert to microns multiply the pixel value by 0.5 um per pixel for z dimension.

X_um, Y_um and Z_um: see description for X_pixels, Y_pixels, and Z_pixels.

Cell_id: Unique identifier for a single cell. “-1“ means that a transcript was detected, but not associated with any segmented cell.

pyxa_studio.csv

Column descriptions:

FOV: The Field Of View (FOV) the cell is in.

Volume_um3: Cell volume based on cell sizes measured in um3.

X_pixels, Y_pixels, and Z_pixels: The x,y and z coordinate (in pixels) of this cell within the FOV, measured in pixels. The pixel edge length is 229.997nm. Thus, to convert to microns multiply the pixel value by 0.22997 um per pixel. For z dimension, the pixel edge length at is 500 nm. Thus, to convert to microns multiply the pixel value by 0.5 um per pixel for z dimension.

Cluster: The cluster ID or numerical label assigned during cell clustering (e.g., from UMAP or Leiden clustering).

Cell_type: cell type annotation assigned to individual cells.

X_UMAP: The UMAP X-axis coordinate — a dimensional reduction value representing where this cell lies in a 2D embedding based on expression similarity.

Y_UMAP: The UMAP Y-axis coordinate — the second dimension of the UMAP embedding. Cells with similar expression profiles appear close together in this space.

segmentation_geometries.parquet

# Load the Parquet file

Open this file with Python via Jupyter Notebook using codes such as:

# import geopandas

import geopandas as gpd

# Load the Parquet file

gdf = gpd.read_parquet("<path/to/parquet.parquet")

# View the first few rows

print(gdf.head())

CellID: same as CellID elsewhere

ZIndex: An index representing the Z-stack plane (or depth) for each cell.

Geometry: The raw geometric data defining the shape of the segmented cell in that specific Z-plane which is used to store spatial data like polygons. Each pair (x, y) defines a point on the polygon boundary (e.g. (4632.265059219087, -20825.509571557806))

'Border': A Boolean value (True or False) indicating whether the segmented cell touches the border of the field of view (FOV).

FOV: field of view (FOV).