template_DE_stats_filename,
local_dir,
num_runs,
project_id,
analysis_type,
col_to_rank_genes,
logFC_name,
pvalue_name,
)
# ### Gene summary table
# In[15]:
summary_gene_ranks = ranking.generate_summary_table(
template_DE_stats_filename, template_DE_stats, simulated_DE_summary_stats,
col_to_rank_genes, local_dir, 'gene', params)
summary_gene_ranks.head()
# In[16]:
# Check if there is an NaN values, there should not be
summary_gene_ranks.isna().any()
# In[17]:
# Create `gene_summary_fielname`
summary_gene_ranks.to_csv(gene_summary_filename, sep='\t')
# ### Compare gene ranking
analysis_type,
col_to_rank_genes,
logFC_name,
pvalue_name,
)
# ## Gene summary table
#
# Note: Using DESeq, genes with NaN in `Adj P-value (Real)` column are those genes flagged because of the `cooksCutoff` parameter. The cook's distance as a diagnostic to tell if a single sample has a count which has a disproportionate impact on the log fold change and p-values. These genes are flagged with an NA in the pvalue and padj columns of the result table. For more information you can read [DESeq FAQs](https://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#pvaluesNA)
# +
summary_gene_ranks = ranking.generate_summary_table(
template_DE_stats_filename,
template_DE_stats,
simulated_DE_summary_stats,
col_to_rank_genes,
local_dir,
"gene",
params,
)
summary_gene_ranks.head()
# -
summary_gene_ranks.isna().any()
# Create `gene_summary_filename`
summary_gene_ranks.to_csv(gene_summary_filename, sep="\t")
# ## Compare gene ranking
# Studies have found that some genes are more likely to be differentially expressed even across a wide range of experimental designs. These *generic genes* are not necessarily specific to the biological process being studied but instead represent a more systematic change.
