orenbenkiki
commented
1 year ago You can only analyze clean cells after creating a clean_gene mask, because part of the computation is taking into account the fraction of these genes UMIs out of the total UMIs.
Edit: It seems you are running v0.8? You should switch to 0.9, and there are vignettes describing the process for it.
SalimMegat
Hi, Sorry but I am still having an issue with picking the right cells. Here the error I get when I run the command : mc.pl.analyze_clean_cells( full, properly_sampled_min_cell_total=PROPERLY_SAMPLED_MIN_CELL_TOTAL, properly_sampled_max_cell_total=PROPERLY_SAMPLED_MAX_CELL_TOTAL, properly_sampled_max_excluded_genes_fraction=PROPERLY_SAMPLED_MAX_EXCLUDED_GENES_FRACTION, )
KeyError Traceback (most recent call last) File:1
File ~/.conda/envs/metacells/lib/python3.8/site-packages/metacells/utilities/logging.py:373, in logged..wrap..wrapper(*args, kwargs)
368 if log_value is not None:
369 logger().log(
370 param_level, "%swith %s: %s", INDENT_SPACES[: 2 INDENT_LEVEL], name, log_value
371 )
--> 373 return function(args, kwargs)
375 finally:
376 if logger().isEnabledFor(step_level):
File ~/.conda/envs/metacells/lib/python3.8/site-packages/metacells/pipeline/clean.py:189, in analyze_clean_cells(adata, what, properly_sampled_min_cell_total, properly_sampled_max_cell_total, properly_sampled_max_excluded_genes_fraction) 187 excluded_adata: Optional[AnnData] = None 188 if properly_sampled_max_excluded_genes_fraction is not None: --> 189 excluded_genes = tl.filter_data(adata, name="dirty_genes", top_level=False, var_masks=["~clean_gene"]) 190 if excluded_genes is not None: 191 excluded_adata = excluded_genes[0]
File ~/.conda/envs/metacells/lib/python3.8/site-packages/metacells/utilities/logging.py:373, in logged..wrap..wrapper(*args, kwargs)
368 if log_value is not None:
369 logger().log(
370 param_level, "%swith %s: %s", INDENT_SPACES[: 2 INDENT_LEVEL], name, log_value
371 )
--> 373 return function(args, kwargs)
375 finally:
376 if logger().isEnabledFor(step_level):
File ~/.conda/envs/metacells/lib/python3.8/site-packages/metacells/tools/filter.py:94, in filter_data(adata, obs_masks, var_masks, mask_obs, mask_var, invert_obs, invert_var, track_obs, track_var, name, top_level) 92 ut.set_o_data(adata, mask_var, var_mask) 93 else: ---> 94 mask = combine_masks(adata, var_masks, invert=invert_var, to=mask_var) 95 if mask is None: 96 assert mask_var is not None
File ~/.conda/envs/metacells/lib/python3.8/site-packages/metacells/utilities/logging.py:373, in logged..wrap..wrapper(*args, kwargs)
368 if log_value is not None:
369 logger().log(
370 param_level, "%swith %s: %s", INDENT_SPACES[: 2 INDENT_LEVEL], name, log_value
371 )
--> 373 return function(args, kwargs)
375 finally:
376 if logger().isEnabledFor(step_level):
File ~/.conda/envs/metacells/lib/python3.8/site-packages/metacells/tools/mask.py:89, in combine_masks(adata, masks, invert, to) 87 else: 88 if must_exist: ---> 89 raise KeyError(f"unknown mask data: {mask_name}") 90 continue 92 if mask.dtype != "bool":
KeyError: 'unknown mask data: clean_gene'
Here the output of mc.pl.analyze_clean_genes : set als.var[properly_sampled_gene]: 17888 true (99.97%) out of 17893 bools set als.var[excluded_gene]: 13 true (0.07265%) out of 17893 bools set als.var[noisy_lonely_gene]: 0 true (0%) out of 17893 bools
Also, I am trying to compute the UMIs of exclude genes in each cells with the function of listed in the notebook but it does not seem to exist in this version of the package ?
Sorry for the harassment,
Many thanks,
Salim.