MULTI-seq
MULTI-seq is a tool for Hashing Demultiplexing created by Seurat. More information can be found at:
MULTI-seq by Seurat: MULTI-seq .
Parameters - Pre processing
These are the same parameters already given for HTODemux, it is not necessary to give them twice for the execution of the pipeline.
| Parameter | Description | Example |
|---|---|---|
| umi | path to filtered rna matrix | –umi /”path/rna_filtered/” |
| hto_matrix | path to filtered hto matrix | –umi “/path/hto_filtered/” |
| selection_method | How to choose top variable features | –selection_method “mean.var.plot” |
| ndelim | delimiter from the cell’s column nam | –ndelim “_” |
| number_features | Number of features to be used when finding variable features | –number_features 2000 |
| assay | Choose assay between RNA or HTO | –assay “HTO” |
| margin | Margin for normalisation | –margin 2 |
| norm_method | Normalisation method | –norm_method “CLR” |
Parameters - Demultiplexing
| Parameter | Description | Example |
|---|---|---|
| quantile_multi | quantile for classification | –quantile_hto 0.07 |
| autoThresh | Whether to perform automated threshold finding to define the best quantile | –autoThresh “FALSE” |
| maxiter | Maximum number of iterations if autoThresh = TRUE | –maxiter 5 |
| qrangeFrom | Lower quantile | –qrangeFrom 0.1 |
| qrangeTo | Upper quantile | –qrangeTo 0.9 |
| qrangeBy | Step to increase/reduce quantile | –qrangeBy 0.0.05 |
The description of this parameters were provided by the library. They can also be found in the link above.