pegasus.arcsinh

pegasus.arcsinh(data, cofactor=5.0, jitter=False, random_state=0, base_matrix=None, target_matrix=None, select=True)[source]

Conduct arcsinh transform on the current matrix.

If jitter == True, jittering by adding a randomized value in U([-0.5, 0.5)). This will also make the matrix dense. Mimic Cytobank.

Parameters

  • cofactor (float, optional, default: 5.0) – Cofactor used in cytobank, arcsinh(x / cofactor).

  • jitter (bool, optional, default: False) – Add a ‘arcsinh.jitter’ matrix in dense format, jittering by adding a randomized value in U([-0.5, 0.5)).

  • random_state (int, optional, default: 0) – Random seed for generating jitters.

  • base_matrix (str, optional, default: None.) – The key name of the matrix to perform arcsinh. If None, the current matrix.

  • target_matrix (str, optional, default: None.) – The key name of the matrix to store the arcsinh results. If None, base_matrix + “.arcsinh”.

  • select (bool, optional, default: None.) – Select the arcsinh matrix as the current matrix (can be accessed via data.X).

Return type

None

Returns

  • None

  • Add the arcsinh matrix to data.matrices.

  • Note that if the detected base_matrix`==`X, we’ll change the name to counts instead.

Examples

>>> pg.arcsinh(data)