2Microhaplotypes are the subject of significant interest in the forensics community as a promising 3 multi-purpose forensic DNA marker for human identification. Microhaplotype markers are 4 composed of multiple SNPs in close proximity, such that a single NGS read can simultaneously 5 genotype the individual SNPs and phase them in aggregate to determine the associated donor 6 haplotype. Abundant throughout the human genome, numerous recent studies have sought 7 to discover and rank microhaplotype markers according to allelic diversity within and among 8 populations. Microhaplotypes provide an appealing alternative to STR markers for human 9 identification and mixture deconvolution, but can also be optimized for ancestry inference or 10 combined with phenotype SNPs for prediction of externally visible characteristics in a multiplex 11 NGS assay. Designing and evaluating panels of microhaplotypes is complicated by the lack of a 12 convenient database of all published data, as well as the lack of population allele frequency data 13 spanning disparate marker collections. We present MicroHapDB, a comprehensive database of 14 published microhaplotype marker and frequency data, as a tool to advance the development of 15 microhaplotype-based human forensics capabilities. We also present population allele frequencies 16 derived from 26 global population samples for all microhaplotype markers published to date, 17 facilitating the design and interpretation of custom multi-source panels. We submit MicroHapDB 18 as a resource for community members engaged in marker discovery, population studies, assay 19 development, and panel and kit design. 20 Well-studied short tandem repeat (STR) markers have formed the basis of forensic human identification 22 methods since the 1990s. The most common strategy in practice today utilizes several fluorescent dyes 23 to type 20 or more STR markers in a single polymerase chain reaction (PCR) followed by capillary 24 electrophoresis (CE) detection (Butler, 2010). The resulting DNA profiles, combined with STR allele 25 frequency estimates, can then be used to calculate match statistics or evaluate the relative weight of evidence 26 1 Standage & Mitchell MicroHapDB v0.5for competing propositions in a likelihood ratio framework (Butler, 2015; Cowell et al., 2015; Bleka et al., 27 2016a,b). Statistics obtained via STR typing can provide high confidence given the number of independent 28 markers in an assay and the multiallelic nature of each marker. 29 Despite impressive recent improvements in DNA sequencing technologies, next-generation sequencing 30 (NGS) assays of single nucleotide polymorphism (SNP) markers have seen slow adoption for forensic 31 human identification. The ability to genotype sufficient numbers of SNPs to achieve suitable statistical 32 power remains beyond the scope of many forensics laboratories. A relevant factor is the forensics 33 community's strong disinclination, on ethical and privacy grounds, to use DNA markers associated 34 with human disease...