Background: for libraries which contain multiplexed samples (for example 10x Genomics CellPlex and Flex data), the reporting currently returns the number and names of the multiplexed samples (derived from the cellranger multi config file) instead of the number and names of the "physical" samples (which correspond to the Fastq files).
In the current implementation the information about multiplexed samples is not explicitly stored anywhere and is instead reported on the fly. However this is a problem for other types of data (specifically at this time Parse Evercode data, but potentially others in future), where the reporting only returns the physical sample information.
As it is not possible at this stage to derive the information for Parse data, an initial proposal to address the problem is:
Create new metadata items to explicitly store the multiplexed sample information
For Parse datasets, set these to non-null placeholder values indicating that the data are missing
Allow these metadata items to be set by explicitly by the user
Use these for reporting, if set to a non-null value
For null values the reporting could fall back to the existing mechanisms (though possibly for 10x Genomics CellPlex and Flex data the metadata should also be explicitly set for completeness e.g. at the time of archiving).
Background: for libraries which contain multiplexed samples (for example 10x Genomics CellPlex and Flex data), the reporting currently returns the number and names of the multiplexed samples (derived from the
cellranger multiconfig file) instead of the number and names of the "physical" samples (which correspond to the Fastq files).In the current implementation the information about multiplexed samples is not explicitly stored anywhere and is instead reported on the fly. However this is a problem for other types of data (specifically at this time Parse Evercode data, but potentially others in future), where the reporting only returns the physical sample information.
As it is not possible at this stage to derive the information for Parse data, an initial proposal to address the problem is:
For null values the reporting could fall back to the existing mechanisms (though possibly for 10x Genomics CellPlex and Flex data the metadata should also be explicitly set for completeness e.g. at the time of archiving).