cellxgene-schema must update validation for X (Matrix Layers)

[brianraymor]

Changelog

X (Matrix Layers)

• Updated Visium Spatial Gene Expression table row to Descendants of Visium Spatial Gene Expression
• Added matrix requirements for Visium CytAssist Spatial Gene Expression, 11mm.

Design

Assay	"raw" required?	"raw" location	"normalized" required?	"normalized" location
Descendant of Visium Spatial Gene Expression	REQUIRED. Values MUST be de-duplicated molecule counts. All non-zero values MUST be positive integers stored as numpy.float32. If uns['spatial']['is_single'] is False then each cell MUST contain at least one non-zero value. If uns['spatial']['is_single'] is True then the unfiltered feature-barcode matrix (raw_feature_bc_matrix) MUST be used. See Space Ranger Feature-Barcode Matrices. if assay_ontology_term_id is "EFO:0022860" for Visium CytAssist Spatial Gene Expression, 11mm, this matrix MUST contain 14336 rows; otherwise, this matrix MUST contain 4992 rows. If the obs['in_tissue'] value is 1, then the cell MUST contain at least one non-zero value. If any obs['in_tissue'] values are 0, then at least one cell corresponding to a obs['in_tissue'] with a value of 0 MUST contain a non-zero value.	AnnData.raw.X unless no "normalized" is provided, then AnnData.X	STRONGLY RECOMMENDED	AnnData.X

[ejmolinelli]

@brianraymor

For this requirement:

If uns['spatial']['is_single'] is True then the unfiltered feature-barcode matrix (raw_feature_bc_matrix) MUST be used. See Space Ranger Feature-Barcode Matrices.

Is this simply a check in the matrix dimensions? How else would we know that the matrix is or is not raw_feature_bc_matrix?

[ejmolinelli]

@brianraymor

When might a cell in a matrix have MORE than one value?

If uns['spatial']['is_single'] is False then each cell MUST contain at least one non-zero value.

[nayib-jose-gloria]

@brianraymor

When might a cell in a matrix have MORE than one value?

If uns['spatial']['is_single'] is False then each cell MUST contain at least one non-zero value.

The "Cell" in this context refers to a biological cell; each row in our matrix represents a cell from our obs dataframe (and each col is a gene). Therefore, we must check that each row in our matrix has at least one non-zero value (that is, ensure each cell has at least one gene with a gene expression value > 0)

[ejmolinelli]

@brianraymor
When might a cell in a matrix have MORE than one value?

If uns['spatial']['is_single'] is False then each cell MUST contain at least one non-zero value.

The "Cell" in this context refers to a biological cell; each row in our matrix represents a cell from our obs dataframe (and each col is a gene). Therefore, we must check that each row in our matrix has at least one non-zero value (that is, ensure each cell has at least one gene with a gene expression value > 0)

Thanks @nayib-jose-gloria , I was thinking of a cell as in a column/row pair in a 2D matrix. This makes a ton of sense!

[nayib-jose-gloria]

Is this simply a check in the matrix dimensions? How else would we know that the matrix is or is not raw_feature_bc_matrix?

correct, all our validator can do is enforce the right matrix dimensions. the rest is instructions for a curator/submitter to do manually