CDR3AAPhyschem

CDR3 AA physicochemical feature analysis

The idea is to perform a regression between two groups of cells (e.g. Treg vs Tconv) at different length of CDR3 AA sequences.
The regression will be performed for each physicochemical feature of the AA (hydrophobicity, volume and isolectric point).

Input

  • scrfile: The data loaded by ScRepCombiningExpression, saved in RDS or qs/qs2 format.
    The data is actually generated by scRepertiore::combineExpression().
    The data must have both TRA and TRB chains.

Output

  • outdir: Default: {{in.immdata | stem}}.cdr3aaphyschem.
    The output directory

Environment Variables

  • group: The key of group in metadata to define the groups to compare. For example, CellType, which has cell types annotated for each cell in the combined object (immdata + Seurat metadata)
  • comparison (type=auto): A dict of two groups, with keys as the group names and values as the group labels. For example, 
    Treg = ["CD4 CTL", "CD4 Naive", "CD4 TCM", "CD4 TEM"]
Tconv = "Tconv"
    Or simply a list of two groups, for example, ["Treg", "Tconv"] when they are both in the group column.
  • target: Which group to use as the target group. The target group will be labeled as 1, and the other group will be labeled as 0 in the regression.
    If not specified, the first group in comparison will be used as the target group.
  • each (auto): A column, or a list of columns or a string of columns separated by comma.
    The columns will be used to split the data into multiple groups and the regression will be applied to each group separately.
    If not provided, all the cells will be used.

Reference