I am currently exploring the integration of multi-modal datasets and was wondering whether LIGER can handle cases with partially overlapping modalities. Specifically, can LIGER integrate datasets structured as follows:
Batch 1: scRNA-seq + scATAC-seq
Batch 2: scATAC-seq only
Batch 3: scRNA-seq only
or cases like:
Batch 1: scRNA-seq + scATAC-seq
Batch 2: scRNA-seq + scATAC-seq
Batch 3: scATAC-seq only
I have carefully reviewed the LIGER tutorials, but could not find a clear answer regarding such scenarios. I would greatly appreciate it if you could provide some guidance on whether LIGER supports this kind of integration, and if so, any recommendations on how to approach it.
I am currently exploring the integration of multi-modal datasets and was wondering whether LIGER can handle cases with partially overlapping modalities. Specifically, can LIGER integrate datasets structured as follows:
Batch 1: scRNA-seq + scATAC-seq
Batch 2: scATAC-seq only
Batch 3: scRNA-seq only
or cases like:
Batch 1: scRNA-seq + scATAC-seq
Batch 2: scRNA-seq + scATAC-seq
Batch 3: scATAC-seq only
I have carefully reviewed the LIGER tutorials, but could not find a clear answer regarding such scenarios. I would greatly appreciate it if you could provide some guidance on whether LIGER supports this kind of integration, and if so, any recommendations on how to approach it.