Scanpy 10x Implementation

[mbcouger]

Hello,

I am trying to use infercnvpy with a 10x dataset. Assuming I follow the tutorial for 10x data and end up with leiden clusters, what would be the commands nescarry to run these on each leiden cluster. This was my best guess which did not work, I beleive I am confused by the syntax.

cnv.pl.chromosome_heatmap(adata, groupby="leiden")
infercnvpy.io.genomic_position_from_gtf('./genes.gtf', adata=None, gtf_gene_id='gene_id', adata_gene_id=None, inplace=True)

What would be the proper way to do this?

Many Thanks!

ErrorLog:

AnnData object with n_obs × n_vars = 30975 × 2886
    obs: 'n_genes', 'n_genes_by_counts', 'total_counts', 'total_counts_mt', 'pct_counts_mt', 'leiden'
    var: 'gene_ids', 'feature_types', 'n_cells', 'mt', 'n_cells_by_counts', 'mean_counts', 'pct_dropout_by_counts', 'total_counts', 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std'
    uns: 'hvg', 'leiden', 'leiden_colors', 'neighbors', 'pca', 'rank_genes_groups', 'umap'
    obsm: 'X_pca', 'X_umap'
    varm: 'PCs'
    obsp: 'connectivities', 'distances'
cnv.pl.chromosome_heatmap(adata, groupby="leiden")
---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-67-f452572c6524> in <module>
----> 1 cnv.pl.chromosome_heatmap(adata, groupby="leiden")

~/anaconda3/lib/python3.8/site-packages/infercnvpy/pl/_chromosome_heatmap.py in chromosome_heatmap(adata, groupby, use_rep, cmap, figsize, show, save, **kwargs)
     56             "'cnv_leiden' is not in `adata.obs`. Did you run `tl.leiden()`?"
     57         )
---> 58     tmp_adata = AnnData(X=adata.obsm["X_cnv"], obs=adata.obs)
     59     chr_pos = list(adata.uns[use_rep]["chr_pos"].values())
     60 

~/anaconda3/lib/python3.8/site-packages/anndata/_core/aligned_mapping.py in __getitem__(self, key)
    146 
    147     def __getitem__(self, key: str) -> V:
--> 148         return self._data[key]
    149 
    150     def __setitem__(self, key: str, value: V):

KeyError: 'X_cnv'

infercnvpy.io.genomic_position_from_gtf('./genes.gtf', adata=None, gtf_gene_id='gene_id', adata_gene_id=None, inplace=True)
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-69-e28d9c88fbac> in <module>
----> 1 infercnvpy.io.genomic_position_from_gtf('./genes.gtf', adata=None, gtf_gene_id='gene_id', adata_gene_id=None, inplace=True)

[Gregor Sturm]

Hi @mbcouger

I think the problem is that you need to run genomic_position_from_gtf first, then tl.infercnv before you can plot anything.
Also you have adata=None for genomic_position_from_gtf, but you need to specify the same AnnData object there you use for the other functions.

Best,
Gregor

[mbcouger]

Hi Gregor,

Thank you I caught this but still have a question on implementation
suppose my gtf is name (genes.gft) What would the command to run

infercnvpy.io.genomic_position_from_gtf('./genes.gtf', adata=None, gtf_gene_id='gene_id', adata_gene_id=None, inplace=True)
I get "name error"

Also if I run the 10x scanpy and get Lendian Clusters, is there a way I can run ithe infercnvpy on a specific cluster?

Many, many thanks for the help!

Cheers,
Brian

[Gregor Sturm]

infercnvpy.io.genomic_position_from_gtf('./genes.gtf', adata=None, gtf_gene_id='gene_id', adata_gene_id=None, inplace=True)
I get "name error"

You still have adata=None here. You need to specify an actual AnnData object.

Also if I run the 10x scanpy and get Lendian Clusters, is there a way I can run ithe infercnvpy on a specific cluster?

You can simply subset adata:

adata = adata[adata.obs["leiden"] == "42", :].copy()

[Gregor Sturm]

I think this has been solved. Feel free to open a new issue if there are other problems.