2.9. ESTIMATE algorithm

ESTIMATE is a well‐established algorithmic tool in the prediction of tumour purity, with ESTIMATE score generated by 141 immune genes and 141 stromal gene expression profiles, using the ESTIMATE R package. Five rounds of gene filtering distinguished the different signatures of m¹A‐, m⁵C‐ and m⁶A‐related regulators: i) a ‘stromal signature’ for the stroma and ii) an ‘immune signature’ for the ICI in tumour tissue. Statistical significance was calculated by integrating the difference between the empirical cumulative distribution function, which is similar to the one used in GSEA, but instead based on absolute expression rather than differential expression.

All analyses were carried out by the R software (version 3.5.2). Distributions of OS were compared using the log‐rank test. The C‐index was used to assess the probability that a prognostic signature with a high value could reflect poor survival. P < 0.05 was considered statistically significant. For multiple comparisons, the Bonferroni corrections were applied following the analysis of variance, and p < 0.05/number of tests served as the significance threshold.