Triplets

Previously established and widely used implementations of ANI begin by either identifying the protein coding genomic fragments^{15 (link)} or extracting approximately 1 Kbp long overlapping fragments^{3 (link)} from the query genome. These fragments are then mapped to the reference genome using BLASTn^{10 (link)} or MUMmer^{30 (link)}, and the best match for each fragment is saved. This is followed by a reverse search, i.e., swapping the reference and query genomes. Mean identity of the reciprocal best matches computed through forward and reverse searches yields the ANI value. Rationale for this bi-directional approach is to bound the ANI computation to orthologous genes and discard the paralogs. In designing FastANI, we followed a similar approach while avoiding the alignment step.
FastANI first fragments the given query genome (A) into non-overlapping fragments of size l. These l-sized fragments are then mapped to the reference genome (B) using Mashmap. Mashmap first indexes the reference genome and subsequently computes mappings as well as alignment identity estimates for each query fragment, one at a time. At the end of the Mashmap run, all the query fragments $f_1,f_2\dots f_{⌊\mid A\mid ∕l⌋}$ are mapped to B. The results are saved in a set M containing triplets of the form 〈f, i, p〉, where f is the fragment id, i is the identity estimate, and p is the starting position where f is mapped to B. The subset of M (say M_forward) corresponding to the maximum identity mapping for each query fragment is then extracted. To further identify the reciprocal matches, each triplet 〈f, i, p〉 in M_forward is ‘‘binned’’ based on its mapping position in the reference, with its value updated to $⟨f,i,bin⟩=⟨f,i,⌊p\mathrm{∕}l⌋⟩$ . Through this step, fragments which are mapped to the same or nearby positions on the reference genome are likely to get equal bin value. Next, M_reciprocal filters the maximum identity mapping for each bin. Finally, FastANI reports the mean identity of all the triplets in M_reciprocal (See Fig. 4 for an example and visualization).Fig. 4

FastANI algorithm explained using synthetic and real examples. a Illustration of FastANI’s work-flow for computing ANI between a query genome and a reference genome. Five mappings are obtained from three query fragments using Mashmap²⁴ . M_forward saves the maximum identity mapping for each query fragment. In this example, M_forward = {m₂, m₄, m₅}. From this set, M_reciprocal picks m₄ and m₅ as the maximum identity mapping for each reference bin. Mapping identities of orthologous mappings, thus found in M_reciprocal, are finally averaged to compute ANI. b FastANI supports visualization of the orthologous mappings M_reciprocal that are used to estimate the ANI value using genoPlotR^{39 (link)}. In this figure, ANI is computed between Bartonella quintana strain (NC_018533.1) as query and Bartonella henselae strain (NC_005956.1) as reference. Red line segments denote the orthologous mappings computed by FastANI for ANI estimation

We define τ as an input parameter to FastANI to indicate a minimum count of reciprocal mappings for the resulting ANI value to be trusted. It is important to appropriately choose the parameters (l, τ, and I₀).

FCS measurements were performed on the Leica TCS SP8 STED 3× microscope (Leica), equipped with a pulsed (80 MHz) white-light laser, HyD detectors, and using a HC PL APO 86× 1.2 NA W motCORR STED (15506333; Leica) water-immersion objective with correction collar. Cells were kept at 37°C during imaging using a Ludin Cube. The microscope was operated with Leice Application Suite, Advanced Fluorescence software in FCS mode. For FCS measurements, the microscope was connected to a PicoHarp 300 stand-alone TCSPC Module (PicoQuant) operated from SymPhoTime 64 software (PicoQuant). We validated the dynamic range of our FCS setup by measuring a dilution series of fluorescein sodium salt (518-47-8; Sigma-Aldrich) in PBS and concluded that we could reliably measure concentrations ranging between 10 nM and 100 µM. For FCS measurements of StableMARK in living cells, U2OS cells were plated on #1.5 18-mm coverslips. The correction collar of the objective was adjusted for every imaged coverslip. 488 nm laser line for excitation and an emission detection window of 500–600 nm were used. Cells were incubated in 10 µM nocodazole for 1 h before the start of FCS measurements. By eye, cells were selected that were classified as low, medium, or high expressing. Per cell, three FCS measurements were performed at different subcellular locations outside the nucleus. The triplet stage model was used to fit the FCS traces and calculate the intracellular concentration of StableMARK.

This study was undertaken with approval from the University of Adelaide Human Ethics Research Committee (H-2013-097 and H-78-2003). Parents provided written informed consent for the use of samples and the data for this research.
Our cohort of 221 Australian children consisted of 108 twin pairs, 2 unpaired twins and one set of triplets identified from the 550 twin/triplet families enrolled in the Tooth Emergence and Oral Health study. Parents/caregivers were required to complete a series of questionnaires as part of the study. The sex of participants was assigned based on parental report. Sex-specific effects were not detected from analysis of ARG diversity or gene abundance. Hence, sex was not included in further analysis of the resistome. A standard medical history was taken at the clinical examination at T3. See Table S5, Supplementary Information for key population characteristics. At T3, the severity of caries was assessed using ICDAS II. In the 66 CA children, the mean ICDAS II score was 1 ± 1.8, while the remaining 145 were caries-free (ICDAS II score = 0).
From the 221 children, 542 oral biofilm samples were initially identified for genomic analysis, of which a total of 12 were excluded, making the final sample size 530. Two were excluded due to antibiotic use (within the past three months). Three were excluded due to extreme sequence depth variation from the mean. This included two with low sequence depth, T1657B_14042014_Q2_Q3 (0.339702 million target reads) and T1472B_25022007_D1_D2 (6.447160 million target reads), and one unusually high sequence depth sample, T1658A_6012009_D1_D2 (117.369 million target reads). We excluded seven samples that contained over 65% host DNA. The eligible samples were from 93 monozygous (MZ), 66 dizygous (DZ), and 59 opposite sex DZ (OSDZ) twins plus and one set of DZ/OSDZ/DZ triplets. One hundred and seventeen children (53%) were sampled at all three time points, 73 (35%) were sampled at 2 time points and 28 (12%) participants sampled at one time point only. While all twins/triplets were samples at the same time, not all individuals had a sample available that met the requirements for stage of dental development for all time points. For this reason, in addition to the post-sequencing removal of 12 samples, there was inconsistent sampling over time.