IndexError: arrays used as indices must be of integer (or boolean) type during Differential Kinetics

[denvercal1234GitHub]

Hi there,

Thanks for the tool.

I was following the Differential Kinetics tutorial and encountered the error below during the scv.tl.differential_kinetic_test. Would you mind helping me to fix it?

Thank you for your help.

My object:

My var_names is a list of 1,285 genes

scv.tl.differential_kinetic_test(adata, var_names=a_list_of_1285_genes, groupby='clusterID')

recovering dynamics (using 1[/10](https://file+.vscode-resource.vscode-cdn.net/10) cores)
100%
1285/1285 [27:02<00:00, 1.23s/gene]
WARNING: CD79B not recoverable due to insufficient samples.
WARNING: S1PR2 not recoverable due to insufficient samples.
WARNING: IFNL1 not recoverable due to insufficient samples.
WARNING: KIR2DL4 not recoverable due to insufficient samples.
WARNING: USP18 not recoverable due to insufficient samples.
    finished (0:27:04) --> added 
    'fit_pars', fitted parameters for splicing dynamics (adata.var)
testing for differential kinetics
... 85%
Output exceeds the [size limit](command:workbench.action.openSettings?%5B%22notebook.output.textLineLimit%22%5D). Open the full output data [in a text editor](command:workbench.action.openLargeOutput?8a675b35-1ae6-4268-b0cb-43bab5891b7c)---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
Cell In[32], line 5
      1 # After basic velocity estimation was done, i.e., scv.pp.filter_and_normalize(adata, min_shared_counts=30, n_top_genes=2000), scv.pp.moments(adata, n_pcs=30, n_neighbors=30), scv.tl.velocity(adata), scv.tl.velocity_graph(adata), scv.pl.velocity_embedding_stream(adata,basis='umap') were ran.
      2 
      3 #var_names = ['Tmsb10', 'Fam155a', 'Hn1', 'Rpl6']
----> 5 scv.tl.differential_kinetic_test(Tonsil_cd8_loomGEX_adata, var_names=Tonsil_cd8_loomGEX_adata_top_likelihoodGenes_Export, groupby='Tissue_Cell_Leiden_clusterID_short')
      8 scv.get_df(Tonsil_cd8_loomGEX_adata[:, Tonsil_cd8_loomGEX_adata_top_likelihoodGenes_Export], ['fit_diff_kinetics', 'fit_pval_kinetics'], precision=2)

File [~/.pyenv/versions/3.10.11/envs/RNAVELO_310/lib/python3.10/site-packages/scvelo/tools/dynamical_model.py:1018](https://file+.vscode-resource.vscode-cdn.net/Users/stillhere/Documents/01_DPhil/scRNAseq/T230T240T246_CXCR5Project/Final_Objects_preAzimuth_2023Feb23/RNA_Velocity_scVelo/RNA_Velocity/SCRIPTS/~/.pyenv/versions/3.10.11/envs/RNAVELO_310/lib/python3.10/site-packages/scvelo/tools/dynamical_model.py:1018), in differential_kinetic_test(data, var_names, groupby, use_raw, return_model, add_key, copy, **kwargs)
   1016 progress = logg.ProgressReporter(len(var_names))
   1017 for i, gene in enumerate(var_names):
-> 1018     dm = DynamicsRecovery(adata, gene, use_raw=use_raw, load_pars=True, max_iter=0)
   1019     if dm.recoverable:
   1020         dm.differential_kinetic_test(clusters, **kwargs)

File [~/.pyenv/versions/3.10.11/envs/RNAVELO_310/lib/python3.10/site-packages/scvelo/tools/dynamical_model.py:22](https://file+.vscode-resource.vscode-cdn.net/Users/stillhere/Documents/01_DPhil/scRNAseq/T230T240T246_CXCR5Project/Final_Objects_preAzimuth_2023Feb23/RNA_Velocity_scVelo/RNA_Velocity/SCRIPTS/~/.pyenv/versions/3.10.11/envs/RNAVELO_310/lib/python3.10/site-packages/scvelo/tools/dynamical_model.py:22), in DynamicsRecovery.__init__(self, adata, gene, load_pars, **kwargs)
     20 super().__init__(adata, gene, **kwargs)
     21 if load_pars and "fit_alpha" in adata.var.keys():
---> 22     self.load_pars(adata, gene)
     23 elif self.recoverable:
     24     self.initialize()

File [~/.pyenv/versions/3.10.11/envs/RNAVELO_310/lib/python3.10/site-packages/scvelo/tools/dynamical_model_utils.py:866](https://file+.vscode-resource.vscode-cdn.net/Users/stillhere/Documents/01_DPhil/scRNAseq/T230T240T246_CXCR5Project/Final_Objects_preAzimuth_2023Feb23/RNA_Velocity_scVelo/RNA_Velocity/SCRIPTS/~/.pyenv/versions/3.10.11/envs/RNAVELO_310/lib/python3.10/site-packages/scvelo/tools/dynamical_model_utils.py:866), in BaseDynamics.load_pars(self, adata, gene)
    863         self.t_ = np.mean(self.t[steady_states])
...
    890     )
--> 891     u, s = u[weights], s[weights]
    892 return u, s

IndexError: arrays used as indices must be of integer (or boolean) type

[qu4drupole]

I ran into the same issue. I'm pretty sure the problem is with the recover dynamics step: WARNING: CD79B not recoverable due to insufficient samples. I was able to resolve the issue by using a more selective list of genes. You could probably just drop those flagged genes, and it would work.