Fig. 1

Experimental design and the integrated analysis of cervical cancer-associated microbiome. (A) Experimental design. A total of 151 human subjects with vaginal metagenomic data were downloaded from NCBI, comprising 42 cervical cancer patients, 43 cervical intraepithelial neoplasia patients, 34 HPV-infected patients and 32 healthy controls. Using MetaPhlan2, the relative abundance of every species in a single sample were calculated. Those species with an average relative abundance greater than 0.5% were selected for further analysis and single nucleotide variant (SNV) calling. All identified SNVs were annotated to their corresponding genes based on their genomic locations. Mutational patterns and signatures were extracted using base substitutions and additionally included information on the sequence context. Subsequently, the top 500 genes exhibiting the highest number of SNVs within each group were selected and annotated to Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolic pathways for assessing functional alterations. Finally, diagnostic models incorporating these multimodal microbial markers and the random forest algorithm was built. (B) Evolutionary tree showing MetaPhlan2 extracts for all species. (C) Abundance of 18 strains selected according to abundance and breadth in different subgroups. (D) Overall screening of and demonstrated that a total of 71,6358 non-redundant mutations were annotated into 25,773 genes belonging to 427 contigs, and the species abundance of the mutations belongs to are also shown