Comparing genomic variant identification protocols for Candida auris

Genomic analyses are widely applied to epidemiological, population genetic and experimental studies of pathogenic fungi. A wide range of methods are employed to carry out these analyses, typically without including controls that gauge the accuracy of variant prediction. The importance of tracking outbreaks at a global scale has raised the urgency of establishing high-accuracy pipelines that generate consistent results between research groups. To evaluate currently employed methods for whole-genome variant detection and elaborate best practices for fungal pathogens, we compared how 14 independent variant calling pipelines performed across 35 Candida auris isolates from 4 distinct clades and evaluated the performance of variant calling, single-nucleotide polymorphism (SNP) counts and phylogenetic inference results. Although these pipelines used different variant callers and filtering criteria, we found high overall agreement of SNPs from each pipeline. This concordance correlated with site quality, as SNPs discovered by a few pipelines tended to show lower mapping quality scores and depth of coverage than those recovered by all pipelines. We observed that the major differences between pipelines were due to variation in read trimming strategies, SNP calling methods and parameters, and downstream filtration criteria. We calculated specificity and sensitivity for each pipeline by aligning three isolates with chromosomal level assemblies and found that the GATK-based pipelines were well balanced between these metrics. Selection of trimming methods had a greater impact on SAMtools-based pipelines than those using GATK. Phylogenetic trees inferred by each pipeline showed high consistency at the clade level, but there was more variability between isolates from a single outbreak, with pipelines that used more stringent cutoffs having lower resolution. This project generated two truth datasets useful for routine benchmarking of C. auris variant calling, a consensus VCF of genotypes discovered by 10 or more pipelines across these 35 diverse isolates and variants for 2 samples identified from whole-genome alignments. This study provides a foundation for evaluating SNP calling pipelines and developing best practices for future fungal genomic studies.

Genomic Epidemiology Linking Nonendemic Coccidioidomycosis to Travel

Coccidioidomycosis is a fungal infection endemic to hot, arid regions of the western United States, northern Mexico, and parts of Central and South America. Sporadic cases outside these regions are likely travel-associated; alternatively, an infection could be acquired in as-yet unidentified newly endemic locales. A previous study of cases in nonendemic regions with patient self-reported travel history suggested that infections were acquired during travel to endemic regions. We sequenced 19 Coccidioides isolates from patients with known travel histories from that earlier investigation and performed phylogenetic analysis to identify the locations of potential source populations. Our results show that those isolates were phylogenetically linked to Coccidioides subpopulations naturally occurring in 1 of the reported travel locales, confirming that these cases were likely acquired during travel to endemic regions. Our findings demonstrate that genomic analysis is a useful tool for investigating travel-related coccidioidomycosis.

One health genomic surveillance and response to a university-based outbreak of the SARS-CoV-2 Delta AY.25 lineage, Arizona, 2021

Genomic surveillance and wastewater tracking strategies were used to strengthen the public health response to an outbreak of the SARS-CoV-2 Delta AY.25 lineage associated with a university campus in Arizona. Epidemiologic and clinical data routinely gathered through contact tracing were matched to SARS-CoV-2 genomes belonging to an outbreak of AY.25 identified through ongoing phylogenomic analyses. Continued phylogenetic analyses were conducted to further describe the AY.25 outbreak. Wastewater collected twice weekly from sites across campus was tested for SARS-CoV-2 by RT-qPCR, and subsequently sequenced to identify variants. The AY.25 outbreak was defined by a single mutation (C18804T) and comprised 379 genomes from SARS-CoV-2 positive cases associated with the university and community. Several undergraduate student gatherings and congregate living settings on campus likely contributed to the rapid spread of COVID-19 across the university with secondary transmission into the community. The clade defining mutation was also found in wastewater samples collected from around student dormitories a week before the semester began, and 9 days before cases were identified. Genomic, epidemiologic, and wastewater surveillance provided evidence that an AY.25 clone was likely imported into the university setting just prior to the onset of the Fall 2021 semester, rapidly spread through a subset of the student population, and then subsequent spillover occurred in the surrounding community. The university and local public health department worked closely together to facilitate timely reporting of cases, identification of close contacts, and other necessary response and mitigation strategies. The emergence of new SARS-CoV-2 variants and potential threat of other infectious disease outbreaks on university campuses presents an opportunity for future comprehensive One Health genomic data driven, targeted interventions.

P484 Expanding VGVI—evidence for distinct Cryptococcus gattii (decagattii) endemic to the American Southwest

 

Poster session 1, September 21, 2022, 12:30 PM - 1:30 PM

Objectives

We aimed to understand the nature of autochthonous Cryptococcus gattii clinical and veterinary cases identified in Arizona, a state in the American Southwest, a locale well outside of the known endemic regions.

Methods

Whole-genome sequencing and phylogenomic comparative analyses were conducted on four unrelated isolates collected from recent cases along with other relevant C. gattii genomes.

Results

Phylogenomic analysis grouped the Arizona genomes with a previously known set of Mexican isolate genomes, labelled as VGVI or C. decagattii. These genomes are clearly delineated from the nearest major molecular type (VGIII), but show no recombination with other molecular types or species of C. gattii.

See Figures below.

Conclusion

These findings expand VGVI into a definitive clade and establish this molecular type as a clinically important and distinct population. These findings also expand the known Cryptococcus ecological range into a previously unrecognized endemic area, typified by extreme heat and aridity.

Molecular type distribution and fluconazole susceptibility of clinical Cryptococcus gattii isolates from South African laboratory-based surveillance, 2005–2013

As is the case globally, Cryptococcus gattii is a less frequent cause of cryptococcosis than Cryptococcus neoformans in South Africa. We performed multilocus sequence typing (MLST) and fluconazole susceptibility testing of 146 isolates randomly selected from 750 South African patients with C. gattii disease identified through enhanced laboratory surveillance, 2005 to 2013. The dominant molecular type was VGIV (101/146, 70%), followed by VGI (40/146, 27%), VGII (3/146, 2%) and VGIII (2/146, 1%). Among the 146 C. gattii isolates, 99 different sequence types (STs) were identified, with ST294 (14/146, 10%) and ST155 (10/146, 7%) being most commonly observed. The fluconazole MIC₅₀ and MIC₉₀ values of 105 (of 146) randomly selected C. gattii isolates were 4 μg/ml and 16 μg/ml, respectively. VGIV isolates had a lower MIC₅₀ value compared to non-VGIV isolates, but these values were within one double-dilution of each other. HIV-seropositive patients had a ten-fold increased adjusted odds of a VGIV infection compared to HIV-seronegative patients, though with small numbers (99/136; 73% vs. 2/10; 20%), the confidence interval (CI) was wide (95% CI: 1.93–55.31, p = 0.006). Whole genome phylogeny of 98 isolates of South Africa’s most prevalent molecular type, VGIV, identified that this molecular type is highly diverse, with two interesting clusters of ten and six closely related isolates being identified, respectively. One of these clusters consisted only of patients from the Mpumalanga Province in South Africa, suggesting a similar environmental source. This study contributed new insights into the global population structure of this important human pathogen.

Cryptococcus is the most common cause of meningitis among adults in South Africa. Most human disease is caused by the members of two species complexes within the genus, Cryptococcus neoformans and Cryptococcus gattii. The environmental range of these species complexes, both found in soil, overlaps in southern Africa though C. gattii is a less common human pathogen. C. gattii is divided into six molecular types: VGI, VGII, VGIII, VGIV, VGV and VGVI. In earlier molecular epidemiology studies including relatively few isolates, most southern African isolates were confirmed as molecular type VGIV. We aimed to determine the molecular diversity of C. gattii in South Africa by genotyping patient isolates obtained through laboratory surveillance, 2005–2013. We confirmed that VGIV was the dominant molecular type and that HIV-seropositive patients were more likely to be infected with VGIV compared to those HIV-seronegative. Analysis of the genomes of South African VGIV isolates revealed that they spanned the whole VGIV clade and confirmed that most isolates did not cluster specifically. However, we observed two interesting clusters of closely related isolates, consisting of patients from three neighbouring provinces in South Africa, suggesting a similar environmental source. Further studies of clinical and environmental African C. gattii isolates are needed to gain a better understanding of this pathogen.

Rhizopus microsporus Infections Associated with Surgical Procedures, Argentina, 2006-2014

Rhizopus spp. fungi are ubiquitous in the environment and a rare but substantial cause of infection in immunosuppressed persons and surgery patients. During 2005–2017, an abnormally high number of Rhizopus infections in surgery patients, with no apparent epidemiologic links, were reported in Argentina. To determine the likelihood of a common source of the cluster, we performed whole-genome sequencing on samples collected during 2006–2014. Most isolates were separated by >60 single-nucleotide polymorphisms, and we found no evidence for recombination or nonneutral mutation accumulation; these findings do not support common source or patient-to-patient transmission. Assembled genomes of most isolates were ≈25 Mbp, and multiple isolates had substantially larger assembled genomes (43–51 Mbp), indicative of infections with strain types that underwent genome expansion. Whole-genome sequencing has become an essential tool for studying epidemiology of fungal infections. Less discriminatory techniques may miss true relationships, possibly resulting in inappropriate attribution of point source.

Genotype, Antifungal Susceptibility, and Virulence of Clinical South African Cryptococcus neoformans Strains from National Surveillance, 2005-2009

In South Africa, Cryptococcus neoformans is the most common cause of adult meningitis. We performed multi locus sequence typing and fluconazole susceptibility testing of clinical C. neoformans isolates collected from 251 South African patients with cryptococcosis through national surveillance from 2005 to 2009. We examined the association between clinical characteristics of patients and genotype, and the effect of genotype on in-hospital mortality. We performed whole genome phylogenetic analysis of fifteen C. neoformans isolates with the molecular type VNB and tested their virulence in a Galleria mellonella model. Most isolates had the molecular type VNI (206/251, 82%), followed by VNII (25/251, 10%), VNB (15/251, 6%), and VNIV (5/251, 2%); 67 sequence types were identified. There were no differences in fluconazole minimum inhibitory concentration (MIC) values among molecular types and the majority of strains had low MIC values (MIC₅₀ of 1 µg/mL and MIC₉₀ of 4 µg/mL). Males were almost twice as likely of being infected with a non-VNI genotype (adjusted odds ratio [OR]: 1.65, 95% confidence interval [CI]: 0.25–10.99; p = 0.61). Compared to patients infected with a VNI genotype, those with a non-VNI genotype had a 50% reduced adjusted odds of dying in hospital (95% CI: 0.03–7.57; p = 0.62). However, for both these analyses, our estimates had wide confidence intervals spanning 1 with large p-values. Fifteen VNB strains were not as virulent in a G. mellonella larval model as the H99 reference strain. A majority of these VNB strains belonged to the VNBII clade and were very closely related by phylogenetic analysis.