Detected Transcripts File Format

The location of decoded MERFISH targets are stored in a csv file called detected_transcripts.csv. All detected_transcripts.csv files have the following columns:

• unlabeled (type: int)

  A numeric index that uniquely identifies a transcript within a field of view. The index is non-consecutive and ascending within each field of view.

• barcode_id (type: int)

  The row index of the identified transcript “barcode” in the codebook file (zero indexed).

• global_x (type: float)

  The x position of the transcript in the global coordinate system, units of microns (um).

• global_y (type: float)

  The y position of the transcript in the global coordinate system, units of microns (um).

• global_z (type: float)

  The index of the z position of the transcript in the experimental image stack.

• x (type: float)

  The x position of the transcript in the field of view (fov) coordinate system, units of pixels.

• y (type: float)

  The y position of the transcript in the field of view (fov) coordinate system, units of pixels.

• fov (type: int)

  The index of the field of view in which the transcript was imaged (zero indexed). fov and barcode_id are a composite primary key for the detected_transcripts.csv table.

• gene (type: string)

  The human readable name of the gene this transcript is associated with. Gene is derived from the “name” column of the codebook file.

• transcript_id (type: int64)

  A unique identifier of the gene that this transcript is associated with. transcript_id is derived from the “id” column of the codebook file.

Following segmentation, the user may choose to append a column to the table to store the EntityID an Entity that contains each transcript. If the transcript is not contained by an Entity, the EntityID will be -1.