NGS provides a large amount of genetic information. Microbial bioinformatics is a data-driven approach that combines the use of sequencing data, machine learning, and artificial intelligence for rapid public health response. This field uses computational tools for disease surveillance, monitoring antimicrobial resistance, and outbreak investigations. By using computer science and statistical methods, such as high-performance supercomputing to organize and interpret the data, bioinformatic tools can track, identify, and monitor pathogens while tracing transmission pathways and phylogenetic origins. Phylogenetic methods play a crucial role in studying the evolutionary history and relationships among organisms. Bioinformatic pipelines are used to assemble genomes, detect genetic variants, and build phylogenies, which are visual representations of the evolutionary relationships among organisms. Those pipelines start with a defined set of files, such as FASTA sequences (a text-based format that represents nucleotide sequences). Connected software routines are then used to generate results, such as sequence alignment or tree figures. Different tools and workflows can also be used to assemble a genome or to perform variant calling (i.e., detecting variants by comparing against a reference genome) (8).

Alignments can identify genetic variations such as single-nucleotide polymorphisms, which are variations in a single nucleotide at a specific place on the genome. Alignments of specific gene sequences or whole genomes aligned to a reference sequence are used as the input for the software or pipelines that generate phylogenies, which trace patterns of shared ancestry among organisms. By analyzing phylogenies, researchers can infer relatedness between pathogen sequences and describe them by using graphics and diagrams such as phylogenetic trees, which illustrate the genetic relationships among organisms. Phylogenetic trees are built by using different probability methods for analysis with various software developed to ensure computational efficiency (8). Phylogenetic trees that show relatedness among pathogens from different sources can provide additional information to complement traditional epidemiology data, determine associations, and help link human cases or establish a common source of infection.

Commercially available bioinformatics pipelines are often used for clinical diagnostic testing. Such pipelines must comply with patient safety, laboratory quality assurance, comparability across laboratories, and local and federal regulatory compliance requirements (9). Many open-source tools, such as Nextstrain (https://www.nextstrain.org), UShER (https://www.genome.ucsc.edu/cgi-bin/hgPhyloPlace), and MicrobeTrace (https://www. microbetrace.cdc.gov), and laboratory-developed code are also available and frequently used. Software containerization methods are used to package bioinformatics tools and pipelines into portable units (or containers), improving efficiency, reproducibility, and security (10), and to combine pathogen genomic information with other sources of information to estimate cases and predict the pathogen’s origins, movements, and potential affect (11).

Scientists have a critical need to share information quickly and efficiently. As a part of the National Institutes of Health, the National Center for Biotechnology Information (NCBI) serves a key role by providing access to biomedical and genomic information, as well as developing software tools for bioinformatics and sequencing analysis. NCBI has served as a hub for sharing genomic information through the GenBank DNA sequence database. The availability of sequences through GenBank enables scientists to compare sequences from other laboratories. AMD is now used to track a wide array of disease agents, including antimicrobial-resistant foodborne bacterial and fungal pathogens, including many in the NCBI Pathogen Detection isolate browser (https://www.ncbi.nlm.nih.gov/pathogens). Other examples of resources that enable sharing of information include the Virus Pathogen Database and Analysis Resource (https://www.bv-brc.org), a platform which provides information about virus mutation, and GISAID (https://www.gisaid.org), an online platform that enables the sharing of information about viral genomic sequences. Those resources play a crucial role in monitoring viral pathogens, including novel strains and respiratory viruses, contributing to the understanding of their evolution and transmission patterns.

Because sequencing costs decreased and platforms were created to manage and analyze larger sets of data, the use of those methods went from proof-of-concept and validating results against traditional epidemiology methods to becoming standard methodologies (12). AMD has become a central part of public health efforts to identify and control infectious diseases and is now incorporated into public health outbreak and emergency response, disease surveillance, drug resistance detection, clinical microbiology, and other public health applications (13).

In the United States, funding provided through CDC has been instrumental in building national capacity for AMD in state, local, and territorial public health laboratories, as well as in hospital and clinical laboratories. The use of AMD has evolved to include a wide array of infectious diseases, including respiratory diseases and antimicrobial drug resistant diseases (14,15). AMD has enhanced public health professionals’ ability to rapidly identify pathogens across the country, track the spread and identify sources of outbreaks, detect drug resistance in US hospitals, inform vaccine development, conduct disease surveillance, and promote international collaborations. Other reports have provided examples of linkages of AMD to surveillance and epidemiology in traveler-based genomic surveillance, enhanced AMR surveillance, and outbreak investigation (16–18). The global relevance of AMD and presence of similar programs in other countries is also of note. For example, a report on the national genomic surveillance system for Listeria monocytogenes and the effect of implementing decentralized sequencing in Australia is included in this issue (19).

The following are a few examples of how AMD is used in the field of public health. First, PulseNet is a national laboratory network that uses AMD diagnostics to detect and prevent foodborne outbreaks (20). PulseNet International involves implementation of whole WGS for global food-borne disease surveillance (21). Second, the MinION portable DNA sequencer was used during the 2014 Ebola virus outbreak in West Africa (22). Third, the Secure HIV TRAnsmission Cluster Engine has been used to identify clusters of highly similar HIV sequences, indicating rapid transmission (23). Fourth, AMD-supported diagnostics were used in the early detection of SARS-CoV-2 variants (24). Finally, rapid AMD-supported diagnostic testing was used during the mpox outbreak response (25).

A necessary part of applied public health is being aware of the caveats and limitations of methods and their applications, as well as understanding questions that public health practitioners should be asking when presented with data derived from NGS methods. Public health practitioners should consider several points when they are given data derived from NGS methods. First, practitioners should ask about the methods used when data are presented and be aware of the limitations of each. Each sequencing methodology has its own limitations. For example, if a lab provides sequencing data using nanopore, there are challenges in using those data for the detection of mutations. Second, practitioners should know a gene being detected doesn’t indicate the gene is being expressed or is functional. Third, practitioners must realize different methods will yield different results; for example, genome assemblies of the same organism by 2 different methods may provide different single-nucleotide polymorphism counts when run on the same panel and could affect epidemiologic investigations; and different bacterial or viral classifiers will yield different results from the same metagenomic raw data. Finally, practitioners should be aware that increasing use of metagenomics presents many unique challenges with results interpretation. When you receive results of interest consider the following: what the method or pipeline was built to do versus what is it doing (i.e., using a SARS-CoV-2 method to look for bacterial DNA); what the result is based on, a single gene or part of a gene; and what databases were used for the analysis. Many pathogens of interest are overrepresented in databases and may turn up disproportionally in results.

Laboratory methods have evolved from relatively simple, culture-dependent techniques for identifying bacteria and viruses to expensive and time-consuming DNA sequencing to more rapid and less costly sequencing, resulting in vast amounts of genetic information. Information about those advances has not always been communicated clearly to healthcare and public health professionals or the public. For readers who are interested in learning more, CDC has enabled state, local, and territorial public health laboratories to provide regional training, including sequencing and molecular epidemiology tools trainings, Bioinformatics Regional Resources, and other opportunities designed to help clinicians, scientists, and public health practitioners to understand transmission chains, characterize emerging pathogens, and solve outbreaks (1) (Table).

Beyond regional trainings, examples of other resources include CDC’s AMD Academy, hosted in partnership with the Association of Public Health Laboratories and the Council of State and Territorial Epidemiologists. The AMD Academy is a multiday training in molecular epidemiology and bioinformatics for epidemiologists and microbiologists from state, territorial, and local health departments. The COVID Genomic Epidemiology toolkit is another valuable resource that addresses topics related to the application of genomics to epidemiologic investigations and public health response to SARS-CoV-2 at state, territorial, and local levels. This toolkit provides introductory information such as how to interpret phylogenetic trees in the context of transmission, along with practical case studies demonstrating real-world applications (26).

In conclusion, this journal supplement aims to improve knowledge and awareness of advances in AMD. The progress of AMD techniques over the past few decades has had a considerable effect on public health. We hope that this article can begin to demystify the field of AMD by providing a brief introduction to other papers in this journal supplement and AMD in general.