The study was approved by the Ethical Committee of the Istanbul University Faculty of Dentistry (No:2014-278) and each patient’s parents provided written informed consent. Children between 3–15 years old without a history of any systemic disease who did not use any medication that reduces saliva flow, or who were not exposed to antimicrobials in the previous three months were included voluntarily to the study.
The teeth were examined under light after dried with cotton rolls. Caries experience was measured using the Decayed, Missing, and Filled Surfaces (DMFS) index26. The DMFS index is applied to the permanent dentition and dmfs index expresses the number of affected tooth surfaces in primary dentition. DMFS ranges zero to 128 with molars and premolars having 5 surfaces and incisors and canines having 4 surfaces. This index accounts for teeth that are restored and missing, and those that are decayed27.
Periodontal health of the participants were assessed using Plaque Index28 (PI) given by Silness and Löe, Gingival Index28 (GI) given by Löe and Silness, and Bleeding on Probing29 (BoP) index given by Ainamo and Bay.
Oral hygiene status was assessed by the PI, the extent of plaque was measured on the buccal surfaces of the teeth and each surface was scored. According to this index, the number of 0 indicates the absence of dental plaque, and the scores between 1–3 indicate the increased presence of dental plaque. Score 0 means no plaque when the explorer is moved across the marginal gingiva. In score 1, the plaque is visible, but when the explorer is moved across the marginal gingiva very little plaques are seen. In score 2, the plaque is visible, the presence of plaque on the tooth surface as a continuous strip along the marginal gingiva. In score 3, the presence of plaque along the marginal gingiva that fills the tooth surface and extends towards the midline, filling the interproximal region28.
Second, the GI was used to evaluate gingival bleeding, which is the main finding of inflammation. GI scores were taken of the mesial, distal, buccal and lingual margins of the tooth. The score from each surface of the tooth was divided by total number of surfaces. According to this index, the number of 0 indicates the absence of gingival inflammation, and the scores between 1–3 indicate the increased presence of gingival inflammation, color change of the gingiva and bleeding. A score of 0 refers to healthy gingiva without signs of inflammation. Score 1 indicates mild discoloration, edema and inflammation, but no spontaneous bleeding or bleeding on probing. Score 2 indicates moderate red discoloration, edema and inflammation, without spontaneous bleeding, but bleeding on probing. In score 3, besides the presence of severe redness, edema and inflammation, spontaneous bleeding is also present28.
For the BoP index, gingival bleeding was recorded as present in a period of 10 seconds after with a periodontal probe. Bleeding areas were recorded as positive (=1), non-bleeding areas were recorded as negative (=0). A total of six measurements around for each tooth teeth were performed. The number of areas with bleeding was divided by the total number of areas examined and multiplied by 100 to score in percent29.
Gender, age, zygosity, type of delivery, and duration of breast-feeding were recorded.
Sample collection and DNA extraction
Each individual was informed to refrain from eating, drinking or cleaning their teeth 2 hours before the examination and sampling. Saliva samples were collected using Saliva DNA Collection and Preservation Devices (Norgen Biotek, CA) according to instructions of the manufacturer. Maximal 2 ml of stimulated saliva samples were collected and preserved at room temperature until DNA extraction. DNA extraction from 500 µl preserved saliva was carried out using Saliva DNA Isolation Kit (Norgen Biotek, CA). Extracted and quantified DNA was stored at −20 °C until further analysis.
Amplification of 16S rRNA gene, library preparation and sequencing
DNA samples were prepared for 300 bp paired end sequencing on MiSeq instrument (Illumina, USA). Each sample was first amplified using dual-indexed fusion primers targeting the hyper variable V3–V4 region of the bacterial 16S rRNA gene. PCR amplification and library preparation was performed using gene specific primers (Bakt_341F: 5′-CCTACGGGNGGCWGCAG-3′ and Bakt_805R: 5′-GACTACHVGGGTATCTAATCC-3′)30 attached to adaptors and multiplex identifier sequences. PCR amplification and library preparation were performed using Phusion High-Fidelity DNA Polymerase. Amplicons were cleaned and normalized using Invitrogen SequalPrep 96-well Plate Kit, then pooled in equimolar ratio and sequenced in a MiSeq instrument (Illumina, USA) with v3 chemistry, which allows 300 bp paired-end reads.
Sequencing data processing
Demultiplexing and clipping of sequence adapters from raw sequences were performed by CASAVA data analysis software (Illumina, USA). The fragments with any mismatches to the barcodes or primers were excluded. After trimming the barcode and primer sequences from the reads, paired-end reads were merged using ‘vsearch join-pairs’ and quality filtered using ‘quality-filter q-score-joined’ within QIIME231. Sequences were denoised using deblur32 with–p-trim-length parameter of 400 and–p-min-reads parameter of 1. Taxonomy was assigned to ASVs using ‘feature-classifier classify-sklearn’ plugin against the pre-trained Naive Bayes classifier (classifier_silva_132_99_16S_V3.V4_341F_805R.qza)33. Bray-Curtis dissimilarity index, weighted and unweighted distance matrixes between twins were generated using ‘diversity core-metrics-phylogenetic’ plugin with–p-sampling-depth parameter of 2,399 to ensure even sampling depth when evaluating diversity.
The study data includes four data types, raw data, metadata (Figshare File F1), ASVs table (Figshare File F2), and ASVs sequences (Figshare File F3)34. Raw sequencing data is available at the National Centre for Biotechnology Information, Sequence Read Archive (SRA)35 while the other data files are available at Figshare34.
Raw sequencing data
The paired-end 16S rRNA sequencing data were deposited in the Sequence Read Archive (SRA) with an accession number PRJNA61358635. The data comprises 396 FASTQ sequence files and there are two files (regarding forward and reverse reads) for each sample.
Samples metadata, ASVs Table and ASVs sequences
Metadata of the studied samples were presented in the Figshare File F134. Metadata information includes sex, age (in years), zygosity status (MZ or DZ), birth type (vaginal or C-section), breast-feeding duration (as months), as well as dental parameters such as caries status, plaque and gingival indexes and bleeding on probing. ASV table (feature table in QIIME2) including read counts per sample is also presented with taxonomy classifications in the Figshare File F234. DNA sequences of ASVs were presented as fasta format in the Figshare File F334.