KeyError: 'gene_symbol' in ppi_deg.plot_network

[sharkLoc]

Please make sure these conditions are met

• I have checked that this issue has not already been reported.
• I have confirmed this bug exists on the latest version of omicverse.
• (optional) I have confirmed this bug exists on the main branch of omicverse.

What happened?

input gene list blow:
['JUN',
'IER2',
'FOS',
'JUNB',
'TEX14',
'DUSP1',
'ZFP36L2',
'RBM38',
'RPS4X',
'FOSB',
'DUSP2',
'DNAJB1',
'TUBA1A',
'CROCC',
'HIST1H4C',
'ZNF331',
'LINC01578',
'LINC00910',
'MGAT4A',
'BTG2',
'C12orf57',
'AL499604.1',
'H3F3B',
'CITED2',
'OSER1',
'BTG1',
'PER1',
'ARRDC2',
'ODC1',
'RPS13',
'CXCR4',
'EIF5A',
'RPL36AL',
'NFKBIA',
'RPL5',
'HNRNPA0',
'EIF1',
'YPEL5',
'RPL3',
'RPL6',
'SNHG5',
'RPL32',
'PNRC1',
'JUND',
'HOOK2',
'TSC22D3',
'SARAF',
'NACA',
'RPS5',
'RPS3',
'LEPROTL1',
'SNHG8',
'KLF2',
'RPS23',
'RPL11',
'SLC2A3',
'SBDS',
'RPL30',
'RPS27A',
'EEF1B2',
'PIK3IP1',
'IER5',
'NPM1',
'ABLIM1',
'SNHG1',
'ATP6V1G1',
'LAPTM5',
'RPL8',
'CIRBP',
'RGCC',
'VAMP2',
'RPS15A',
'ITM2A',
'EEF1A1',
'RPL29',
'RPL9',
'RPL7',
'AIF1',
'COX4I1',
'RPL24',
'TBCC',
'EIF2S3',
'PHB2',
'RPL14',
'TNFAIP3',
'TOMM7',
'PFDN5',
'APOO',
'RPL18A',
'EEF1D',
'RPL35A',
'RBM3',
'GADD45B',
'H1FX',
'RPL19',
'KDM6B',
'RPS14',
'RPL22',
'ZBTB10',
'AC022217.3']

Minimal code sample

def run_ppi(df,group, topn=50):
    gene = df['gene'].head(topn)
    ppi_deg = ov.bulk.pyPPI(gene= gene, 
                        gene_type_dict= dict(zip(gene, ['up']*topn)), 
                        gene_color_dict=  dict(zip(gene, ['#4DBBD5FF']*topn)), 
                        species=9606)
    ppi_deg.interaction_analysis()

    fig2, axes2 = ppi_deg.plot_network(
            figsize = (12,12), 
            plot_node_num = topn,
            pos_type = 'kamada_kawai', #'kamada_kawai',
            node_size= 20
    )
    fig2.savefig("{}_gene_ppi.pdf".format(group), bbox_inches='tight', dpi=300)

Error output

KeyError                                  Traceback (most recent call last)
Cell In[45], line 1
----> 1 run_ppi(df7,"CD4_T_naive_Ab_vs_Ag_PBMC")

Cell In[42], line 9, in run_ppi(df, group, topn)
      3 ppi_deg = ov.bulk.pyPPI(gene= gene, 
      4                     gene_type_dict= dict(zip(gene, ['up']*topn)), 
      5                     gene_color_dict=  dict(zip(gene, ['#4DBBD5FF']*topn)), 
      6                     species=9606)
      7 ppi_deg.interaction_analysis()
----> 9 fig2, axes2 = ppi_deg.plot_network(
     10         figsize = (12,12), 
     11         plot_node_num = topn,
     12         pos_type = 'kamada_kawai', #'kamada_kawai',
     13         node_size= 20
     14 )
     15 fig2.savefig("{}_gene_ppi.pdf".format(group), bbox_inches='tight', dpi=300)

File /mnt/p0/biotools/python-3.10.12/lib/python3.10/site-packages/omicverse/bulk/_network.py:220, in pyPPI.plot_network(self, **kwargs)
    210 def plot_network(self,**kwargs) -> Tuple[matplotlib.figure.Figure,matplotlib.axes._axes.Axes]:
    211     r"""Plot protein-protein interaction network.
    212 
    213     Arguments:
   (...)
    218         ax: Axes object containing PPI network plot
    219     """
--> 220     return plot_network(self.G,self.gene_type_dict,self.gene_color_dict,**kwargs)

File /mnt/p0/biotools/python-3.10.12/lib/python3.10/site-packages/omicverse/utils/_plot.py:639, in plot_network(G, G_type_dict, G_color_dict, pos_type, pos_dim, figsize, pos_scale, pos_k, pos_alpha, node_size, node_alpha, node_linewidths, plot_node, plot_node_num, label_verticalalignment, label_fontsize, label_fontfamily, label_fontweight, label_bbox, legend_bbox, legend_ncol, legend_fontsize, legend_fontweight)
    636     pos=nx.kamada_kawai_layout(G,dim=pos_dim,scale=pos_scale)
    637 degree_dict = dict(G.degree(G.nodes()))
--> 639 G_color_dict=dict(zip(G.nodes,[G_color_dict[i] for i in G.nodes]))
    640 G_type_dict=dict(zip(G.nodes,[G_type_dict[i] for i in G.nodes]))
    642 nx.draw_networkx_edges(G, pos,nodelist=list(G_color_dict.keys()), alpha=pos_alpha)

File /mnt/p0/biotools/python-3.10.12/lib/python3.10/site-packages/omicverse/utils/_plot.py:639, in <listcomp>(.0)
    636     pos=nx.kamada_kawai_layout(G,dim=pos_dim,scale=pos_scale)
    637 degree_dict = dict(G.degree(G.nodes()))
--> 639 G_color_dict=dict(zip(G.nodes,[G_color_dict[i] for i in G.nodes]))
    640 G_type_dict=dict(zip(G.nodes,[G_type_dict[i] for i in G.nodes]))
    642 nx.draw_networkx_edges(G, pos,nodelist=list(G_color_dict.keys()), alpha=pos_alpha)

KeyError: 'H4C6'

Versions

Details

🔬 Starting plot initialization...
🧬 Detecting GPU devices…
✅ NVIDIA CUDA GPUs detected: 1
    • [CUDA 0] NVIDIA RTX A5500
      Memory: 22.1 GB | Compute: 8.6

   ____            _     _    __                  
  / __ \____ ___  (_)___| |  / /__  _____________ 
 / / / / __ `__ \/ / ___/ | / / _ \/ ___/ ___/ _ \ 
/ /_/ / / / / / / / /__ | |/ /  __/ /  (__  )  __/ 
\____/_/ /_/ /_/_/\___/ |___/\___/_/  /____/\___/                                              

🔖 Version: 1.7.8   📚 Tutorials: https://omicverse.readthedocs.io/
✅ plot_set complete.

[Starlitnightly]

@claude

[claude]

Claude finished @Starlitnightly's task in 2m 18s —— View job • claude/issue-500-20260211-0622 • Create PR ➔

---

KeyError in ppi_deg.plot_network - Fixed

• Read CLAUDE.md for repository guidelines
• Examine the source code to understand the bug
• Identify root cause of gene symbol mismatch
• Implement a fix
• Commit and push changes
• Provide PR link

Root Cause

The issue occurs because STRING database returns standardized gene symbols that may differ from the input gene names. For example:

• Input: HIST1H4C
• STRING returns: H4C6 (the preferred name)

When plot_network tries to create color and type dictionaries for the graph nodes (line 645-646 in omicverse/utils/_plot.py), it fails because:

1. The graph nodes use STRING's preferred names (H4C6)
2. But gene_color_dict and gene_type_dict still use the original input names (HIST1H4C)

This causes a KeyError when trying to look up the node name in the original dictionaries.

Solution Implemented

I've made two changes to fix this issue:

1. Updated pyPPI.interaction_analysis() in omicverse/bulk/_network.py:

   • Added a call to string_map() to get the mapping from input gene names to STRING preferred names
   • Updated gene_type_dict and gene_color_dict to use STRING preferred names as keys
   • This ensures the dictionaries match the actual node names in the graph

2. Added safeguard in plot_network() in omicverse/utils/_plot.py:

   • Changed dictionary lookups to use .get() with defaults (#808080 for color, 'unknown' for type)
   • This handles cases where STRING adds interacting partner genes that weren't in the original input list

These changes ensure that the gene symbols in the dictionaries always match the node names in the PPI network graph.

Testing

The fix should resolve the KeyError: 'H4C6' error. The code will now:

• Correctly map input gene names like HIST1H4C to their STRING preferred names like H4C6
• Handle any additional interacting partners added by STRING that weren't in the original input

---

|