Table 3.

Seven ASVs that showed significant agreement* with regards to the probabilities of presence or absence (Kappa statistics) between saliva and any pancreatic tissue or intestinal samples

ASV IDb	Kappa (SD)	q-value	Agree obs	Prop.pp	Prop.pa	Prop.ap	Prop.aa	Genusa	Species (HOMD)a
ASV28	0.406 (0.116)	0.004	0.896	0.042	0.104	0.000	0.854	Fusobacterium	nucleatum subsp. vincentii
ASV23	0.492 (0.137)	0.004	0.896	0.063	0.083	0.021	0.833	Rothia	mucilaginosa (GG)
ASV20	0.431 (0.142)	0.013	0.875	0.063	0.083	0.042	0.813	Saccharibacteria_(TM7)_[G-1]	bacterium_HMT_352
ASV27	0.357 (0.111)	0.009	0.875	0.042	0.125	0.000	0.833	Oribacterium	parvum
ASV01	0.344 (0.119)	0.016	0.708	0.563	0.271	0.021	0.146	Streptococcus
ASV47	0.309 (0.104)	0.014	0.917	0.021	0.000	0.083	0.896	Fusobacterium	nucleatum subsp. vincentii
ASV17	0.284 (0.101)	0.017	0.729	0.083	0.271	0.000	0.646	Fusobacterium

Agree obs = observed agreement; ASV = Amplicon Sequence Variants; SD = standard deviation; q-value = p-value adjusted for multiple testing via the false discovery rate method; Prop. pp = probability of an ASV to be present in both oral and any pancreatic tissue or intestinal samples; Prop. pa = probability of an ASV to be present in oral samples but absent in pancreatic or intestinal samples; Prop. ap = probability of an ASV to be absent in oral samples but present in pancreatic or intestinal samples; Prop. aa = probability of an ASV to be absent in both oral and any pancreatic tissue or intestinal samples.

*Significant agreement/concordance can be reached if probability of an ASV to be present in both oral and any pancreatic tissue or intestinal samples (Prop. pp), or probability of an ASV to be absent in both oral and any pancreatic tissue or intestinal sample (Prop. aa) is high.

^aTAXONOMIC annotations of the ASVs are from HOMD database unless otherwise noted. GG = green gene database.

^bSV ID used in Figure 1 left panel.