Table 1.
Data are for 154 adolescents and 31,571 adults allocated into ranking groups by their intake of total milk, i.e., non-fermented plus fermented milk.
Fig 1.
The plots show mean taxa abundances for the lowest milk intake group on the x-axis and the mean abundance difference between the lowest and highest tertile groups on the y-axis for A) saliva and B) tooth biofilm.
Fig 2.
Participant clustering and saliva taxa associated with milk intake.
(A) PLS separation of participants based on saliva microbiota characterization by Illumina MiSeq sequencing displayed in a PLS score scatter plot. The scores t1 and t2 on the x- and y-axes are the new created variables summarizing the x variables. Red dots indicate adolescents in the lowest tertile of total milk intake, and blue dots those in the highest tertile. The two lower bar graphs list taxa with statistically significant correlation or a PLS correlation coefficient ≥0.1 for (B) being in the highest tertile of total milk intake and (C) in the lowest tertile. Colored bars are for taxa with statistically significant PLS correlation coefficients, i.e. the 95% CI does not include zero.
Fig 3.
Participant clustering and tooth biofilm taxa associated with milk intake.
(A) PLS separation of participants based on tooth biofilm microbiota characterization by Illumina MiSeq sequencing displayed in a PLS score scatter plot. The scores t1 and t2 on the x- and y-axes are the new created variables summarizing the x variables. Red dots indicate adolescents in the lowest tertile of total milk intake, and blue dots those in the highest tertile. The two lower bar graphs list taxa with statistically significant correlation or a PLS correlation coefficient ≥0.1 for (B) being in the highest tertile of total milk intake and (C) in the lowest tertile. Colored bars are for taxa with statistically significant PLS correlation coefficients, i.e. the 95% CI does not include zero.
Fig 4.
Box plot from mutans streptococci and lactobacilli culture.
The plot illustrates colony forming units (CFU) per ml of whole chewing-stimulated saliva of (A) mutans streptococci, and (B) lactobacilli. Data are from culturing on MSB and Rogosa agar, respectively and are expressed as 10log counts.
Table 2.
Detection of mutans streptococci by cultivation and DNA based methods in adolescents classified into tertiles based on their reported intake of total milk, i.e., non-fermented milk plus fermented milk.
Fig 5.
The plot illustrates the association of self-reported sweet and non-sweet snack intake from the questionnaire, BMI, and makers of S. mutans and lactobacilli in tooth and saliva samples in a PLS model where tertile group allocation from total milk intake were the dependent variables. Variables to the left (blue) were associated with being in the highest tertile and those to the right (red) with being in the lowest tertile. The relative importance of each x-variable in the projection is expressed by a variable importance in the projection (VIP). VIP-value ≥1.0 are considered influential and ≥1.5 as highly influential. VIP-values are listed to the right.
Table 3.
Reported intake of snack products in 154 adolescents and 31,571 adults.
Data are presented as means with 95% confidence limits adjusted for sex and BMI in adolescents and sex, age, BMI, and education in adults.
Fig 6.
The plot illustrates the associations between milk intake levels (low, middle, and high tertiles) and saliva taxa differing significantly by milk intake in PLS (cf Figs 1 and 2), frequency of sweet snacks, and PCR detection of S. mutans (S. m) and cultivation of mutans streptococci (log CFU ms). Taxa are scaled and listed by relative abundance with those associated with low milk intake highlighted in red and those associated with high intake in blue. The relative abundancies are illustrated by the sizes of each colour segment in the outer circle. The percentages outside the milk group segments refer to the contribution from each of the variables but variable labels are omitted.
Fig 7.
Circos plots for tooth biofilm taxa.
The plot illustrates the associations between milk intake levels (low, middle, and high tertiles) and for tooth biofilm taxa differing significantly by milk intake in PLS (cf Figs 1 and 2), frequency of sweet snacks, and PCR detection of S. mutans (S. m) and cultivation of mutans streptococci (log CFU ms). Taxa are scaled and listed by relative abundance with those associated with low milk intake highlighted in red and those associated with high intake in blue. The relative abundancies are illustrated by the sizes of each colour segment in the outer circle. The percentages outside the milk group segments refer to the contribution from each of the variables but variable labels are omitted.