The SNPforID browser: an online tool for query and display of frequency data from the SNPforID project

The SNPforID browser is a web-based tool for the query and visualization of the SNP allele frequency data generated by the SNPforID consortium ( http://www.snpforid.org/ ). From this project, validated panels of single nucleotide polymorphisms (SNPs) for a variety of forensic applications have been generated with the browser concentrating on the single-tube identification SNP set comprising 52 markers. A web interface allows the visitor to review the allele frequencies of the studied markers from all the available populations used by SNPforID to validate global SNP variability. The interface has been designed to offer the useful facility of combining populations into appropriate geographic groups for visual comparison of populations individually or amongst user-defined groupings and with equivalent HapMap data.     

Relationship between the 1359 G/A polymorphism of the Central Cannabinoid Receptor 1 (CNR1) gene and susceptibility to cannabis addiction in a Turkish population

In this study, we investigate the existence of a possible genetic association between 1359 G/A polymorphism of the Central Cannabinoid Receptor 1 (CNR1) Gene CNR1 (p.Thr453Thr; rs1049353) single nucleotide polymorphism (SNP) and cannabis addiction. DNA samples used in this work are purified from venous leukocytes of 145 unrelated Turkish cannabis-dependent subjects and 140 Turkish control subjects. No significant difference is observed in genotype or allele frequencies of CNR1 1359 G/A polymorphism between these two groups. We also compared CNR1 1359 G/A polymorphism allele frequency distribution in our healthy Turkish population with other healthy populations. The comparison of healthy Turkish subjects with the healthy subjects from English-Irish, Chinese, European-American, African-American, Italian, German and Japanese populations revealed significant differences in allele frequencies. Data indicate that the 1359 G/A CNR1 polymorphism does not contribute to susceptibility to cannabinoid addiction in Turkish subjects. To the best of our knowledge, this is the first study on 1359 G/A CNR1 polymorphism in the Turkish population.     

Establishing a second-tier panel of 18 ancestry informative markers to improve ancestry distinctions among Asian populations

At present, several mature ancestry informative SNP (AISNP) panels are used to distinguish between continental regions of the world, but a more accurate division within the continent requires a secondary panel to complete. However, many AISNPs for the subgroup ancestry inference are selected from the Kidd Lab panel of 55 AISNPs or other published papers. These panels inevitably lack valuable markers for subgroup ancestry inference. Therefore, instead of choosing from the published panels, we used the 1000 Genomes Project to screen potentially informational markers in Asian populations, including single nucleotide polymorphisms (SNPs) and insertion/deletion polymorphisms (InDels). The allele frequencies of all autosomal SNPs and InDels of the 1000 Genomes Project were compared between 10 populations in Asia to identify markers with the largest pairwise allele frequency differences. Finally, we established a second-tier panel of 18 AIMs in this study, which not only divided the 26 populations of the 1000 Genomes Project into six clusters, but also divided the Asia subgroup into four clusters: Gujarati, East Asia, Southeast Asia and South Asia.     

SNPSTR rs59186128_D7S820 polymorphism distribution in European Caucasoid, Hispanic, and Afro-American populations

Single nucleotide polymorphisms (SNPs) in the flanking regions of microsatellite loci (SNPSTRs) help to increase the power of discrimination of short tandem repeat (STR) loci. SNPs are positions in the genome that have been well-conserved over the course of evolution, so analysing them can help distinguish between STR alleles in which the number of repetitions matches due to descent from those which match by chance. This provides support for the determination of biological paternity and other kinship analyses in which mutation needs to be ruled out as grounds for exclusion. Locus D7S820 shows a variable position, SNP rs59186128, in the 5′ flanking region. This study is set out (1) to determine the frequencies of SNP rs59186128 in populations with various geographical origins and (2) to estimate the possible contribution of rs59186128 to the allele discrimination of locus D7S820. To that end, individuals from European Caucasoid, Hispanic, and Afro-American populations are studied using denaturing high-performance liquid chromatography, which enables locus rs59186128 to be quickly and highly cost-effectively screened. Moreover, a method is established for determining the haplotypes of SNPSTR rs59186128_D7820. The results show that SNP rs59186128 has a T allele frequency of more than 0.15 in one of the Afro-American populations studied, and the haplotype analysis shows that there is no preferential association between the alleles of SNPSTR rs59186128_D7S820, which supports the idea that they could be useful in forensic applications.     

A hypervariable STR polymorphism in the CFI gene: Southern origin of East Asian-specific group H alleles

Previous studies of four populations revealed that a hypervariable short tandem repeat (iSTR) in intron 7 of the human complement factor I (CFI) gene on chromosome 4q was unique, with 17 possible East Asian-specific group H alleles observed at relatively high frequencies. To develop a deeper anthropological and forensic understanding of iSTR, 1161 additional individuals from 11 Asian populations were investigated. Group H alleles of iSTR and c.1217A allele of a SNP in exon 11 of the CFI gene were associated with each other and were almost entirely confined to East Asian populations. Han Chinese in Changsha, southern China, showed the highest frequency for East Asian-specific group H alleles (0.201) among 15 populations. Group H alleles were observed to decrease gradually from south to north in 11 East Asian populations. This expansion of group H alleles provides evidence that southern China and Southeast Asia are a hotspot of Asian diversity and a genetic reservoir of Asians after they entered East Asia. The expected heterozygosity values of iSTR ranged from 0.927 in Thais to 0.874 in Oroqens, higher than those of an STR in the fibrinogen alpha chain (FGA) gene on chromosome 4q. Thus, iSTR is a useful marker for anthropological and forensic genetics.     

An overview of SNP-SNP microhaplotypes in the 26 populations of the 1000 Genomes Project

Microhaplotypes (MHs) are a promising new type of forensic markers that are defined by the combinations of two- or more single-nucleotide polymorphisms (SNPs) within 200 bp. Their advantages, such as low mutation rates, lack of stutter artifacts, and short amplicons, have improved human identification, kinship analysis, ancestry prediction, and mixture deconvolution capabilities. Information on published MHs, e.g., allele frequencies, is available in widely used public databases, ALlele FREquency Database, and MicroHapDB. However, there are abundant non-published MHs spread over the whole genome, and those databases do not incorporate other databases (e.g., the SNP Database) to provide users with more integrated information. Therefore, it is essential to establish a robust, responsive, and comprehensive MHs database. In this study, we thoroughly screened for SNP-SNP MHs among 26 populations from the 1000 Genomes Project (Phase 3). All genotype data of SNPs in each MH were converted to PHASE input files, and allele frequencies were estimated using PHASE. We compiled a detailed summary of SNP-SNPs at the global, continental, and population levels focused on haplotypes and the A_e value and supplemented our database using dbSNP data (last updated in 2015). We have successfully established a dual-SNP MH database (D-SNPsDB) of MHs within 50 bp for 26 populations in the integration of basic data such as physical positions in the human genome, mapping of variant identifiers (rsIDs), allele frequencies, and basic variant information. For public database queries, the D-SNPsDB web app was developed with the R Shiny package to get integrated information.

     

Forensic identity SNPs: Characterisation of flanking region variation using massively parallel sequencing

Single nucleotide polymorphisms (SNPs) can be analysed for identity or kinship applications in forensic genetics to either provide an adjunct to traditional STR typing or as a stand-alone approach. The advent of massively parallel sequencing technology (MPS) has provided a useful opportunity to more easily deploy SNP typing in a forensic context, given the ability to simultaneously amplify a large number of markers. Furthermore, MPS also provides valuable sequence data for the targeted regions, which enables the detection of any additional variation seen in the flanking regions of amplicons. In this study we genotyped 977 samples across five UK-relevant population groups (White British, East Asian, South Asian, North-East African and West African) for 94 identity-informative SNP markers using the ForenSeq DNA Signature Prep Kit. Examination of flanking region variation allowed for the identification of 158 additional alleles across all populations studied. Here we present allele frequencies for all 94 identity-informative SNPs, both including and excluding the flanking region sequence of these markers. We also present information on the configuration of these SNPs in the ForenSeq DNA Signature Prep Kit, including performance metrics for the markers and investigation of bioinformatic and chemistry-based discordances. Overall, the inclusion of flanking region variation in the analysing workflow for these markers reduced the average combined match probability 2175 times across all populations, with a maximum reduction of 675,000-fold in the West African population. The gain due to flanking region-based discrimination increased the heterozygosity of some loci above that of some of the least useful forensic STR loci; thus demonstrating the benefit of enhanced analysis of currently targeted SNP markers for forensic applications.     

Gender is a major factor explaining discrepancies in eye colour prediction based on HERC2/OCA2 genotype and the IrisPlex model

In recent years, several studies have greatly increased our understanding of the genetic basis underlying human eye colour variation. A large percentage of the eye colour diversity present in humans can already be genetically explained, so much so that different DNA-based eye colour prediction models, such as IrisPlex, have been recently developed for forensic purposes. Though these models are already highly accurate, they are by no means perfect, with many genotype-phenotype discrepancies still remaining unresolved. In this work we have genotyped six SNPs associated with eye colour (IrisPlex) in 535 individuals from Spain, a Mediterranean population. Aside from different SNP frequencies in Spain compared to Northern Europe, the results for eye colour prediction are quite similar to other studies. However, we have found an association between gender and eye colour prediction. When comparing similar eye colour genetic profiles, females tend, as a whole, to have darker eyes than males (and, conversely, males lighter than females). These results are also corroborated by the revision and meta-analysis of data from previously published eye colour genetic studies in several Caucasian populations, which significantly support the fact that males are more likely to have blue eyes than females, while females tend to show higher frequencies of green and brown eyes than males. This significant gender difference would suggest that there is an as yet unidentified gender-related factor contributing to human eye colour variation.     

Further development of forensic eye color predictive tests

In forensic analysis predictive tests for external visible characteristics (or EVCs), including inference of iris color, represent a potentially useful tool to guide criminal investigations. Two recent studies, both focused on forensic testing, have analyzed single nucleotide polymorphism (SNP) genotypes underlying common eye color variation (Mengel-From et al., Forensic Sci. Int. Genet. 4:323 and Walsh et al., Forensic Sci. Int. Genet. 5:170). Each study arrived at different recommendations for eye color predictive tests aiming to type the most closely associated SNPs, although both confirmed rs12913832 in HERC2 as the key predictor, widely recognized as the most strongly associated marker with blue and brown iris colors. Differences between these two studies in identification of other eye color predictors may partly arise from varying approaches to assigning phenotypes, notably those not unequivocally blue or dark brown and therefore occupying an intermediate iris color continuum. We have developed two single base extension assays typing 37 SNPs in pigmentation-associated genes to study SNP-genotype based prediction of eye, skin, and hair color variation. These assays were used to test the performance of different sets of eye color predictors in 416 subjects from six populations of north and south Europe. The presence of a complex and continuous range of intermediate phenotypes distinct from blue and brown eye colors was confirmed by establishing eye color populations compared to genetic clusters defined using Structure software. Our study explored the effect of an expanded SNP combination beyond six markers has on the ability to predict eye color in a forensic test without extending the SNP assay excessively – thus maintaining a balance between the test's predictive value and an ability to reliably type challenging DNA with a multiplex of manageable size. Our evaluation used AUC analysis (area under the receiver operating characteristic curves) and naïve Bayesian likelihood-based classification approaches. To provide flexibility in SNP-based eye color predictive tests in forensic applications we modified an online Bayesian classifier, originally developed for genetic ancestry analysis, to provide a straightforward system to assign eye color likelihoods from a SNP profile combining additional informative markers from the predictors analyzed by our study plus those of Walsh and Mengel-From. Two advantages of the online classifier is the ability to submit incomplete SNP profiles, a common occurrence when typing challenging DNA, and the ability to handle physically linked SNPs showing independent effect, by allowing the user to input frequencies from SNP pairs or larger combinations. This system was used to include the submission of frequency data for the SNP pair rs12913832 and rs1129038: indicated by our study to be the two SNPs most closely associated to eye color.     

Eurasiaplex-2: Shifting the focus to SNPs with high population specificity increases the power of forensic ancestry marker sets

To compile a new South Asian-informative panel of forensic ancestry SNPs, we changed the strategy for selecting the most powerful markers for this purpose by targeting polymorphisms with near absolute specificity – when the South Asian-informative allele identified is absent from all other populations or present at frequencies below 0.001 (one in a thousand). More than 120 candidate SNPs were identified from 1000 Genomes datasets satisfying an allele frequency screen of ≥ 0.1 (10 % or more) allele frequency in South Asians, and ≤ 0.001 (0.1 % or less) in African, East Asian, and European populations. From the candidate pool of markers, a final panel of 36 SNPs, widely distributed across most autosomes, were selected that had allele frequencies in the five 1000 Genomes South Asian populations ranging from 0.4 to 0.15. Slightly lower average allele frequencies, but consistent patterns of informativeness were observed in gnomAD South Asian datasets used to validate the 1000 Genomes variant annotations. We named the panel of 36 South Asian-specific SNPs Eurasiaplex-2, and the informativeness of the panel was evaluated by compiling worldwide population data from 4097 samples in four genome variation databases that largely complement the global sampling of 1000 Genomes. Consistent patterns of allele frequency distribution, which were specific to South Asia, were observed in all populations in, or closely sited to, the Indian sub-continent. Pakistani populations from the HGDP-CEPH panel had markedly lower allele frequencies, highlighting the need to develop a statistical system to evaluate the ancestry inference value of counting the number of population-specific alleles present in an individual.     

Two deoxyribonuclease I gene polymorphisms and correlation between genotype and its activity in Japanese population

Deoxyribonuclease I (DNase I) plays important roles for DNA fragmentation and degradation during programmed cell death. The single nucleotide polymorphism (SNP) at DNase I, designated as DNASE1, in exon 8 (A2317G) is considered to be one of the susceptibility genes for gastric and colorectal carcinoma and myocardial infarction. Recent research has shown the presence of a novel 56-bp variable number of tandem repeat (VNTR) polymorphism in intron 4 at DNase I, designated as HumDN1. In the present study, DNASE1 and HumDN1polymorphisms and serum DNase I activities in each different genotype were investigated in 137 Japanese populations. The allele frequencies of A and G in DNASE1 were 0.5839 and 0.4161, respectively. The allelic frequencies of alleles 2, 3, 4, and 5 in HumDN1 were 0.0219, 0.5803, 0.2226, and 0.1752, respectively. In the DNASE1 polymorphism, the activities of genotypes GG and AG were significantly higher than that of AA. As for the HumDN1 polymorphism, the activity of genotype 55 was significantly higher than the activities of 33 and 34. In addition, a significant difference was observed between haplotypes AA/33 and GG/55. The analysis of the correlation between genotype and DNase I activity may be potentially useful for clinical purposes.     

Comparison of manual and automated AmpliSeq™ workflows in the typing of a Somali population with the Precision ID Identity Panel

The Precision ID Identity Panel was used to type 109 Somali individuals in order to obtain allele frequencies for the Somali population. These frequencies were used to establish a Somali HID-SNP database, which will be used for the biostatistic calculations in family and immigration cases. Genotypes obtained with the Precision ID Identity Panel were found to be almost in complete concordance with genotypes obtained with the SNPforID PCR-SBE-CE assay. In seven SNP loci, silent alleles were identified, of which most were previously described in the literature.The project also set out to compare different AmpliSeq™ workflows to investigate the possibility of using automated library building in forensic genetic case work. In order to do so, the SNP typing of the Somalis was performed using three different workflows: 1) manual library building and sequencing on the Ion PGM™, 2) automated library building using the Biomek^®3000 and sequencing on the Ion PGM™, and 3) automated library building using the Ion Chef™ and sequencing on the Ion S5™. AmpliSeq™ workflows were compared based on coverage, locus balance, noise, and heterozygote balance. Overall, the Ion Chef™/Ion S5™ workflow was found to give the best results and required least hands-on time in the laboratory. However, the Ion Chef™/Ion S5™ workflow was also the most expensive. The number of libraries that may be constructed in one Ion Chef™ library building run was limited to eight, which is too little for high throughput workflows. The Biomek^®3000/Ion PGM™ workflow was found to perform similarly to the manual/Ion PGM™ workflow. This argues for the use of automated library building in forensic genetic case work. Automated library building decreases the workload of the laboratory staff, decreases the risk of pipetting errors, and simplifies the daily workflow in forensic genetic laboratories.     

Revision of the SNPforID 34-plex forensic ancestry test: Assay enhancements,standard reference sample genotypes and extended population studies

A revision of an established 34 SNP forensic ancestry test has been made by swapping the under-performing rs727811 component SNP with the highly informative rs3827760 that shows a near-fixed East Asian specific allele. We collated SNP variability data for the revised SNP set in 66 reference populations from 1000 Genomes and HGDP-CEPH panels and used this as reference data to analyse four U.S. populations showing a range of admixture patterns. The U.S. Hispanics sample in particular displayed heterogeneous values of co-ancestry between European, Native American and African contributors, likely to reflect in part, the way this disparate group is defined using cultural as well as population genetic parameters. The genotyping of over 700 U.S. population samples also provided the opportunity to thoroughly gauge peak mobility variation and peak height ratios observed from routine use of the single base extension chemistry of the 34-plex test. Finally, the genotyping of the widely used DNA profiling Standard Reference Material samples plus other control DNAs completes the audit of the 34-plex assay to allow forensic practitioners to apply this test more readily in their own laboratories.     

Allele frequencies of three STRs of the human von Willebrand factor gene (vWF) in a Brazilian population sample

Allele frequencies were calculated for three tetrameric short tandem repeats (STRs) located in intron 40 of the human von Willebrand factor (vWA, vWF1 and vWF2) in 352 white individuals sampled from an urban population from the northeastern region of the State of São Paulo, Brazil. The exact test did not indicate any significant deviation from HWE for any of the three investigated loci. The allele frequencies of vWA and vWF1 showed unimodal and bimodal distributions, respectively, and the frequencies of vWF2 in our sample exhibited bimodal or even trimodal patterns. These differing patterns could reflect the differential action of one selective factor or of the distribution of mutations in these STRs, although the STRs are very close to one another and belong to the same gene. The frequency of paternity exclusions observed for each of these three loci conform to the theorectical expectations. The lack of difficulties regarding the methodology of typing and the forensic value of statistical parameters confirm the usefulness of these systems to study Brazilian populations.     

Increasing the reference populations for the 55 AISNP panel: the need and benefits

Ancestry inference for an individual can only be as good as the reference populations with allele frequency data on the SNPs being used. If the most relevant ancestral population(s) does not have data available for the SNPs studied, then analyses based on DNA evidence may indicate a quite distantly related population, albeit one among the more closely related of the existing reference populations. We have added reference population allele frequencies for 14 additional population samples (with >1100 individuals studied) to the 125 population samples previously published for the Kidd Lab 55 AISNP panel. Allele frequencies are now publicly available for all 55 SNPs in ALFRED and FROG-kb for a total of 139 population samples. This Kidd Lab panel of 55 ancestry informative SNPs has been incorporated in commercial kits by both ThermoFisher Scientific and Illumina for massively parallel sequencing. Researchers employing those kits will find the enhanced set of reference populations useful.

     

Typing of Amerindian Kichwas and Mestizos from Ecuador with the SNPforID multiplex

A total of 119 unrelated individuals from two of the major ethnic groups in Ecuador were typed for 49 of the autosomal single nucleotide polymorphisms (SNPs) in the SNPforID 52plex using the SNapShot(?) assay. Of the above, 42 samples originated from Mestizos (an admixed population) and the remaining 77 were from Native Amerindian Kichwas. We obtained full SNP profiles in all individuals and concordance of duplicated analyses. No deviation from Hardy-Weinberg equilibrium (HWE) was observed for any SNP in the Mestizo and Kichwa populations and only one and four pairs of loci, respectively showed significant linkage disequilibrium. A relatively low genetic diversity and global positive F(IS) value was observed in Kichwas. A statistically significant global F(ST) value was obtained when the two Ecuadorian populations were compared with populations in Spain, Portugal, Argentina, Denmark, Greenland, China, Somalia and Mozambique. All pairwise F(ST) values were statistically significant. A multi-dimensional scaling based on pairwise F(ST) values showed that the Kichwa population differed from all other populations investigated and that the Mestizos had an intermediate position between Kichwas and Europeans. An admixture analysis indicated that the greater contributor to the Mestizo population was the Kichwas (71.2%) compared to the European contribution. The combined mean match probability and mean paternity exclusion probability were 3.3 × 10(-17) and 0.998, respectively, for the Mestizo population and 3.3 × 10(-14) and 0.993, respectively, for the Kichwa population.     

Population genetic analyses of 15 STR loci from seven forensically-relevant populations residing in the state of Kuwait

Allele frequencies and population statistics of 15 forensically-important STR loci were estimated for seven populations residing in the State of Kuwait. The populations were: Saudi Arabian, Iraqi, Egyptian, Iranian, Sri Lankan, Bangladeshi, and Indian. All loci were highly polymorphic. After correction, only one locus in one population sample (Saudi Arabian) showed departure from Hardy–Weinberg equilibrium. The populations cluster (MDS Plot) generally with geographic distance and substructure effects for even these seven diverse populations is Fst < 0.01.     

A single nucleotide polymorphism panel for individual identification and ancestry assignment in Caucasians and four East and Southeast Asian populations using a machine learning classifier

Single nucleotide polymorphism (SNP) profiling is an effective means of individual identification and ancestry inferences in forensic genetics. This study established a SNP panel for the simultaneous individual identification and ancestry assignment of Caucasian and four East and Southeast Asian populations. We analyzed 220 SNPs (125 autosomal, 17 X-chromosomal, 30 Y-chromosomal, and 48 mitochondrial SNPs) of the DNA samples from 563 unrelated individuals of five populations (89 Caucasian, 234 Taiwanese Han, 90 Filipino, 79 Indonesian and 71 Vietnamese) and 18 degraded DNA samples. Informativeness for assignment (In) was used to select ancestry informative SNPs (AISNPs). A machine learning classifier, support vector machine (SVM), was used for ancestry assignment. Of the 220 SNPs, 62 were individual identification SNPs (IISNPs) (51 autosomal and 11 X-chromosomal SNPs) and 191 were AISNPs (100 autosomal, 13 X-chromosomal, 30 Y-chromosomal, and 48 mitochondrial SNPs). The 51 autosomal IISNPs offered cumulative random match probabilities (cRMPs) ranging from 1.56 × 10−21 to 3.16 × 10−22 among these five populations. Using AISNPs with the SVM, the overall accuracy rate of ancestry inference achieved in the testing dataset between Caucasian, Taiwanese Han, and Filipino populations was 88.9%, whereas it was 70.0% between Caucasians and each of the four East and Southeast Asian populations. For the 18 degraded DNA samples with incomplete profiling, the accuracy rate of ancestry assignment was 94.4%. We have developed a 220-SNP panel for simultaneous individual identification and ethnic origin differentiation between Caucasian and the four East and Southeast Asian populations. This SNP panel may assist with DNA analysis of forensic casework.     

On the use of dense sets of SNP markers and their potential in relationship inference

With the advent of high density SNP arrays and the progress of next generation sequencing, demands for new methods to handle the subsequent data analysis have exploded. Forensic laboratories are generally hesitant to implement new methods in casework unless they are thoroughly tested and validated. This is particularly true when a third party contractor is involved in the analysis. In this paper we explore data from dense sets of SNP markers and study how different errors could potentially affect the results. Particularly, we study the effects of genotyping errors, linkage disequilibrium as well as the use of inappropriate population frequencies. We demonstrate that ignoring these concepts may lead to false conclusions for some different relationship cases and outline solutions to mitigate these problems.     

Comparison of the genetic background of different Colombian populations using the SNPforID 52plex identification panel

Various strategies for analysing SNP markers and genotyping have been published with the goal of obtaining informative profiles from biological samples that contain only small amounts of template and/or degraded DNA. In this study, a multiplex assay of 52 autosomal single-nucleotide polymorphisms (SNPs) was used to analyse 438 individuals from urban populations from different regions of Colombia, as well as a sample of 50 Native American individuals of the Pastos ethnic group from Nariño. To determine if significant differences in these 52 SNPs exist between the distinct regions of Colombia, genetic distance and admixture analyses were performed based on the available data for 17 different Colombian population groups and for population groups from Africa, Europe and America. The results demonstrate significant differences between the populations from the Southwest Andean, Central-West Andean, Central-East Andean, Orinoquian and northern Colombian Pacific Coast regions. Most of the regions exhibited a European and Native American admixture. One exception is the population from the region of Chocó (on the northern Pacific Coast), which exhibits a high proportion of African admixture (54 %). From the observed genetic backgrounds, it is possible to conclude that a single reference database for the entire country would not be suitable for forensic purposes. The allele frequencies and the forensically relevant parameters were calculated for all of the markers in each Colombian region with significant values for the combined matching probability (power of discrimination ≥0.99999999999999990) and the combined probability of exclusion (≥0.9990) in trios that were obtained from all of the population groups.