https://orcid.org/0000-0002-3120-0174
Figures
Abstract
Molecular source tracking (MST) can improve community health by enabling the identification of the source species of fecal bacteria contamination in waterways. However, widespread adoption of this method at a large scale is hindered by the cost of commercial extraction kits and the technical expertise required to use them. We developed a simpler, efficient, scalable, accessible, and semi-quantitative method to extract DNA from environmental water samples via heat lysis. After filtering water samples onto a polycarbonate membrane, the membrane is suspended in AE buffer, heated, and then centrifuged. The liquid supernatant is then used directly in quantitative PCR (qPCR) analysis. Our filter-heat-spin (FHS) extraction method was compared to a commercial DNA extraction kit (Qiagen DNeasy). In qPCR the traditional kit extraction yielded 2.4 times more copies/μL (on average) of the target sequence, but the FHS extraction yielded 5.1 times more DNA (on average) than the traditional extraction kit due to the higher volume of buffer used in the FHS extraction. Additionally, FHS maintained identical long-term stability as compared to the kit extraction over a five-week period. Further, the cost per FHS extraction is roughly $0.05 per sample and requires 15 minutes to complete, while the typical kit extraction is roughly $4.48 per sample and requires 50 minutes to complete. The proposed FHS extraction method allows for efficient and inexpensive water sample processing, which in turn, reduces barriers for the implementation of MST.
Author summary
The study developed a simple DNA extraction method for use on environmental freshwater samples. Water samples collected from sample sites are filtered through a 0.22 micron pore size polycarbonate membrane trapping bacteria and other cells on the membrane. Next the membrane is suspended in AE (aqueous elution) buffer. The sample is then heated to 95oC for 10 minutes to lyse cells and release the DNA into the solution. To obtain the supernatant with the DNA, it is centrifuged. The samples can then be used in Molecular Source Tracking via qPCR to quantify bacteria levels or bacteria species sources in waterways. When compared to commercially available kits, this filter-heat-spin (FHS) method costs significantly less money, has fewer steps that are less prone to pipetting errors, and is quicker. The DNA has a similar yield and stability. The FHS DNA extraction method can help make Molecular Source Tracking of bacteria more viable for municipalities and researchers, especially in cases with a large number of sample sites.
Citation: Pham VT, Harris-Ryden Z, Kodali NS, Hamner TF, Popalzai S, Hernandez MC, et al. (2025) Filter, heat, spin: A simple and inexpensive method for DNA preparation from freshwater for use in high-throughput molecular source tracking. PLOS Water 4(8): e0000293. https://doi.org/10.1371/journal.pwat.0000293
Editor: Guillaume Wright, PLOS: Public Library of Science, UNITED KINGDOM OF GREAT BRITAIN AND NORTHERN IRELAND
Received: August 30, 2024; Accepted: July 10, 2025; Published: August 12, 2025
Copyright: © 2025 Pham et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data used to produce this article are included in Supporting Information.
Funding: Funds were provided by the University of Texas at Austin, College of Natural Sciences instructional budget (https://cns.utexas.edu/) (SR; VP; ZHR; NK; TH; SP; MCH; MA; MO; TR). Texas Institute for Discovery Education in Science - San Antonio Livestock Exposition research award (https://www.sarodeo.com/p/about/scholarship) (VP) and the Freshman Research Initiative summer fellowship (https://fri.cns.utexas.edu) (VP). Texas Institute for Discovery Education in Science TEJAS fellowship (https://cns.utexas.edu/student-experience/learn-doing) (VP; SP). National Science Foundation S-STEM fellowship (https://new.nsf.gov/funding/opportunities/s-stem-nsf-scholarships-science-technology-engineering-mathematics) (MCH). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Aging infrastructure can pose risks including exposure to potentially pathogenic microbes, harmful chemicals, and heavy metals [1–4]. There is a correlation between these risks and Fecal Indicator Bacteria (FIB) levels making this a good index for the health of an aquatic system [5]. This data can inform management decisions as well as infrastructure planning [5–7]. Molecular Source Tracking (MST) is an even more powerful technique as it allows for the identification of the FIB source species (e.g., human, dog, other domesticated animals) in creeks, lakes, and waterways [8]. The biological source and quantity of FIB can point to the underlying pollution source (e.g., runoff, leaching pipes) which, in turn, allows watershed managers to better develop possible solutions to protect public health and environmental quality.
There are a multitude of studies that utilize the MST of FIB to detect nonpoint sources of pollution for the purposes of infrastructure planning and public health management. [9–11]. A common assay used for the detection and quantification of human-associated FIB in freshwater samples is based on amplifying the HF183 16S ribosomal DNA cluster which is specific to human-associated FIB [12–15]. While the knowledge gained from molecular assays is highly useful, they typically require extensive sample preparation that includes the extraction and purification of DNA from water samples using molecular biology kits (such as the Qiagen DNeasy Blood and Tissue Kit or DNeasy PowerWater Kit) [15]. These kits utilize various buffer solutions to resuspend bacteria, lyse cells, bind DNA to a matrix, wash away any contaminating proteins and/or biomolecules, and finally, elute the highly pure DNA off the matrix for further analysis. However, these commercial molecular biology kits are expensive, require trained technicians, and are time intensive. In fact, sample preparation via traditional extraction methods for a qPCR assay accounts for approximately half the total cost of MST [16].
A simpler, less expensive, and faster method of sample preparation for MST analysis would eliminate the need for a commercial DNA extraction kit, thus reducing costs and preparation time. Instead of relying on traditionally used lysis buffers, the proposed method uses heat to lyse the microbes and inactivate proteins to isolate DNA from water samples for the HF183 TaqMan qPCR assay protocol. Heat lysis has been shown to be an effective method of bacterial DNA extraction from blood culture fluid; for this application, it is also more sensitive, economical, and higher throughput than traditional methods [17]. The advantages of FHS include lower cost, higher efficiency, and less time. Labs will be able to dramatically increase throughput, thus, allowing for more widespread adoption of MST. For example, high throughput sampling via FHS can be advantageous when tracing FIBs [11,18,19] or disease outbreaks such as COVID-19 and polio in wastewater [20–24]. The goal of this project was to determine if the FHS method could give comparable yields of DNA from environmental freshwater samples versus a commonly used kit. A DNA preparation method that is simpler, less expensive, and faster has allowed our lab to perform MST from several samples in one day.
Materials and methods
Sample collection
Water samples were obtained from 15 separate collections along Waller Creek within the University of Texas at Austin main campus via previously described methods [15]. Samples were collected from two locations, one denoted SB and the other CS. Collection dates were: SB 1–3 July 23, 2019; SB 4–6 and CS 10–12 January 27, 2022; and SB 7–9 and CS 13–15 June 2, 2022. Waller Creek is known to have a high amount of fecal bacteria, particularly human associated fecal bacteria [25]. Before collecting the samples, 500 mL HDPE plastic amber bottles were acid washed with 10% hydrochloric acid. At the site, the bottles were rinsed with creek water 3–5 times before collecting approximately 500 mL of creek water in each bottle. The bottles were then stored for no longer than two hours on ice in an insulated container to prevent exposure to sunlight and high temperatures.
Filtration
Filter towers were placed on a filter manifold using Nalgene analytical filter funnels equipped with 0.22 micron pore size polycarbonate filter membranes. Following the filtration of 200 mL samples, filters were folded and placed in 1.5 mL microcentrifuge tubes. The filters extracted with the DNeasy kit had acid washed glass beads added while the filters extracted by the FHS method were resuspended in Qiagen AE Buffer.
Commercial extraction kit
The Qiagen DNeasy blood and tissue kit (Qiagen catalog #69504) is one of the currently accepted standard methods for extracting DNA from freshwater samples. This kit was chosen because it has been used for years [16,26], and it is the least expensive kit for DNA extraction from water (Qiagen.com). Other studies have also found little to no differences between the Qiagen DNeasy blood and tissue kit and other specialty kits [26,27]. We followed standard manufacturer protocols for performing DNA extractions with the recommended U.S. EPA modifications. In brief, the microcentrifuge tube containing the filter and glass beads was resuspended in 500 μL of ATL buffer and 40 μL of proteinase K. The sample was then run in a bead beater for 2 minutes before being centrifuged at 12,000 x g for 2 minutes. Next, the liquid contained within the microcentrifuge tube was transferred into a new 2 mL microcentrifuge tube before being centrifuged at 12,000 x g for 1 minute. Then 225 μL of liquid was transferred into another new microcentrifuge tube and the standard protocol for the Qiagen DNeasy Blood and Tissue Kit was followed. The eluted solution was collected in a microcentrifuge tube and stored in a -20˚C freezer. This method yields approximately 100 μL of starting DNA template for qPCR.
FHS only
As shown in Fig 1, the FHS method begins by resuspending the filter in 500 μL AE buffer (10 mM Tris-Cl and 0.5 mM EDTA, pH 9.0). After vortexing the microcentrifuge tube for 10 seconds, it was then heated for 10 minutes at 95°C in a heat block. This solution was then vortexed again for 10 seconds and centrifuged at 12,000 x g for 2 minutes. Following centrifugation, the liquid was decanted into a separate microcentrifuge tube. This method yields approximately 450 μL of starting DNA template for qPCR which was distributed among aliquots and frozen at -20°C.
qPCR
The qPCR was performed using the well established HF183 sequence for detecting human-derived Bacteroides [13]. The qPCR reaction was run on a BioRad 96-Well CFX Connect Real-Time PCR Detection System using the standard protocol for the New England Biolabs LUNA 2X probe-based qPCR mastermix, with 40 cycles. The cycles were as follows: 95˚C for 10 minutes for activation followed by 40 cycles of 95˚C for 15 seconds and 60˚C for 60 seconds (followed by a plate read). Reactions were run in a total volume of 20 µL with 10 µL of LUNA master mix, 1.66 µL forward primer, 1.66 µL reverse primer (0.83 µM final concentration), 1.32 µL probe (0.26 µM final concentration), 0.36 µL of molecular biology grade water, and 5 µL of either sample, gBlock dilutions (as the positive control), or molecular biology grade water (to serve as no template controls). The sequences for the forward and reverse primers along with the probe are in Table 1, and Table 2 shows the HF183 gBlock that was used to produce the standard curve of 102, 103, 104 copies/µL. These concentrations were used to make the standard curve as prior literature has shown that the range of HF183 in water samples is typically within that range [10]. The standard curve was created by plotting the cycle threshold vs the log of the standards, and performing linear regression only using standard curves with an R2 value greater than 0.9. Using the standard curves, the qPCR efficiency was calculated as follows- Fig 2 and 3 equals 91%, Fig 4 equals 94%, and Fig 5 equals 104%. A slope-intercept formula was generated, allowing for the calculation of the starting quantity (SQ) from the relative cycle threshold (CQ). Since the samples were run in quadruplicates, the average SQ was used to represent the final value for the sample. Outliers were removed from the average SQ calculation by testing relative CQ’s for normality using the Shapiro-Wilks test, then using the Grubbs test or IQR method depending on normality. Each qPCR plate also included an internal HF183 standard to normalize runs across different plates.
The numbered abbreviations each represent the location and sample number of the 15 different collections along Waller Creek in Austin, TX at two different sites (SB and CS). The asterisks represent a statistical significance, p < 0.05, between the commercial kit and FHS method in 6 out of the 15 sample sites.
Total copy number was calculated by multiplying the copy number/µL by the total volume of the sample. The final volume of the FHS extraction method is 450 µL while for the commercial kit it is 100 µL. Only in sample SB03 was the FHS method total yield of DNA statistically higher, p < 0.05, denoted by the asterisk. The numbered abbreviations each represent the location and sample number of the 15 different collections along Waller Creek in Austin, TX collected at two different sites (SB and CS).
Each data point represents six simultaneous water collections. Within each collection there are four technical replicates. Error bars represent the standard error of the mean. A Student’s t-test was used to compare the SQ values each week, which showed no statistical significance, p > 0.05.
Eight samples were collected along the creek on July 22, 2021 and processed the same day. The E. coli was quantified using 3M Petrifilm plates and ranged from 300 to 2033 cfu/100mL. The HF183 range was2809 to 14403 copy number/100mL (Copy number/µL was converted to copy number/100mL for easier comparison to FIB).
Time course study
Six samples were collected at two sites along Waller Creek and extracted using both the commercial kit and FHS method. Following extraction, the samples were aliquoted and stored at -20oC. The qPCR assay outlined above was performed at one week intervals over the course of five weeks.
Statistical analysis
All qPCR reactions were run in 4 technical replicates per sample. Quadruplicate Cq values were averaged, if only three out of four replicates amplified, then an average Cq was calculated with the fourth replicate assigned a Cq value of 40. Cq averages within the range of quantification of the master standard curve spanning 3 orders of magnitude were converted to starting quantity (copies/µL). Non-detects were assigned to samples if three or more replicates did not amplify within 40 cycles or if the calculated concentration was below the limit of detection of the qPCR assay. GraphPad Prism Version 10.2.3 was used to generate figures and perform statistical analyses.
Results
The copy number/µL for each extraction method at two sampling sites over three different days is shown in Fig 2. When comparing the average copy number/µL between the commercial extraction method and the FHS method, in 6 of the 15 samples, the commercial extraction method yielded a statistically higher copy number/µL as determined by the Student’s T-test, p < 0.05 (SB04, SB05, SB06, CS10, CS11, and CS12). For the other 9 of 15 samples (SB01, SB02, SB03, SB07, SB08, SB09, CS13, CS14, and CS15) there was no statistical difference in copy number/µL. Fig 3 shows the total quantity of DNA calculated by multiplying the copy number/µL by the total volume of the sample. The final volume of the FHS extraction method is 450 µL while for the commercial kit it is 100 µL. In 14 of the 15 samples, no statistically significant difference in total DNA was seen between the methods. Only in sample SB03 was the FHS method total yield of DNA statistically higher, p < 0.05. So while the commercial kit produces a more concentrated sample than the FHS method, because of the higher volume of the FHS sample, the total quantity of DNA is about the same. Furthermore, both methods gave reliable results with very low variability within each extraction method. It should be noted that the variance in Fig 2 and 3 between trials can be attributed to the samples being collected on different days (see methods). While all of the samples were collected at base flow (>7 days post-rain), FIB levels vary due to season and climatological conditions.
Since the FHS method does not involve any purification of the DNA, the next step of the investigation aimed to determine whether the presence of impurities in the FHS extracted samples would lead to increased DNA degradation over time. The relative stability of the extraction methods was tested over a period of 5 weeks. DNA extracted from each method was aliquoted and stored at -20°C. One vial was thawed each week and used in qPCR targeting the HF183 sequence. The calculated copy number/µL for each week was compared to the initial copy number/µL. As seen in Fig 4, there was no statistically significant degradation in either method.
As a proof of concept, the FHS method was applied to samples collected at 8 different sites along Waller Creek in Austin, TX. Fig 5 shows the copy number/100mL of HF183 DNA and FIB colony forming units per 100 mL (cfu/100mL) at 8 different sites along the length of the creek. Because the FIB levels measure total E. coli and HF183 detects human-derived Bacteroides, the quantities are not directly comparable. As we found the trend in HF183 copies does not directly correlate with E. coli levels, however, this does demonstrate that FHS works to detect HF183 in water samples containing FIB. As other studies have shown [28,29] the lack of a direct trend here suggests that other species sources (such as canine or avian) are also responsible for the FIB contamination.
Discussion
The FHS DNA extraction protocol was developed to test the viability of a high-throughput method for quickly and inexpensively extracting DNA from freshwater samples. Across 15 independent samples from two different sampling sites and three different dates, the FHS extraction method yields viable DNA that is equally stable when compared to the commercial kit extraction method. The commercial kit yields a volume of 100 µL, while the FHS method yields a volume of 450 µL. The concentration of the commercial extraction method is sometimes higher, but the total HF183 copies produced with the FHS method is the same or greater (Fig 2 and 3). The variability between the different samples is due to the different locations and different dates that the samples were collected.
The lack of purification steps in the FHS method compared to the commercial kit extraction could lead to the presence of nucleases and PCR inhibitors in samples that might potentially alter their stabilities or impair quantification. However, as seen in Fig 4, the DNA extracted by both methods had similar stabilities over a period of five weeks. The results of this experiment suggest that the high temperature (95˚C) in the FHS extraction was enough to inactivate enzymes that could lead to DNA degradation. However, it should be noted that the presence of inhibitors in the template solution could potentially disrupt DNA detection, and diluting the template may be necessary to overcome such issues [26; 30,31].
The FHS method is faster to perform which allows it to be higher throughput. Vastly decreased sample processing time makes MST a more accessible method for monitoring FIB source species. The commercial kit also requires six different pipetting steps while the FHS extraction method only requires two. Fewer steps means that the FHS method has a lower chance of introducing variability between samples which, in turn, allows for personnel with less training to be able to successfully extract DNA with reliable yields. Tracking FIB contamination in watersheds necessitates a large number of data points across both time and space. The time and resources needed to use a standard DNA extraction kit make it difficult to achieve the high throughput that is necessary for such large-scale projects (Table 3). The FHS method solves many of these problems, and as seen in Fig 5, performing MST across a watershed becomes possible. Our group has been studying urban creek FIB levels in Austin, Texas, and to date have processed over 400 samples using the FHS method.
The ability for high-throughput processing of water samples for qPCR analysis can be useful for a range of studies. For example, while the City of Austin monitors the FIB levels in Austin’s creeks to identify areas of high contamination, they have not identified the species source of the FIB contamination [32]. Pinpointing the sources of FIB contamination requires the processing of many samples at the same time (we commonly sample from 8-15 sites in duplicate). Additionally, [33] have used FHS to perform rapid detection of Enterococci by both qPCR and Loop-mediated isothermal amplification. Other examples are investigations of how rainfall affects FIB levels across multiple regions [18,19]. High-throughput sampling can also be advantageous when tracing disease outbreaks such as COVID-19 and polio in wastewater [20–24]. Our lab has also used the FHS extraction with TaqMan DogBac, used for identifying canine associated fecal indicator bacteria [34,35]. Running DogBac assays alongside HF183 allows for comparison of FIB sources. This method could also be applied to detect bovine and pig specific fecal bacteria markers, expanding its applicability from urban environments to rural nonpoint molecular source tracking [34,36].
Conclusion
The FHS method offers several advantages that allow for the rapid, scalable, and semi-quantitative extraction of DNA for use in qPCR. The speed and decreased costs of the FHS method make it practical to answer experimental questions that were previously difficult and costly to answer. We have demonstrated the usefulness of this method in our own research group mapping the species source of FIB contamination in waterways across the city of Austin, TX. These creek-wide collections typically involve sampling from 8 to 15 sites in a single morning, and FHS makes the MST of so many samples time and cost effective. While FHS does not purify the DNA, we have shown that this cruder extraction is sufficient for DNA amplification based detection of FIB, but it has limitations. For example we found that it did not work for preparing creek samples for microbiome DNA sequencing. Nonetheless, solutions to high FIB in waterways require a fuller understanding of the sources of contamination that FHS makes possible due to its significantly lower cost and quicker processing of the samples. Additionally qPCR can be used to detect potential pathogens and antibiotic resistance genes. We hope that the FHS procedure will make MST more accessible to researchers and municipalities that will lead to deeper understanding of the microbes in waterways and solutions to limit harm to aquatic ecosystems.
Supporting information
S1 Table. Sample Sites (Fig 2 & 3): Information for the water sample sites used in figures 2 and 3 including: the date collected, longitude and latitude, days since last precipitation, and flow rate.
https://doi.org/10.1371/journal.pwat.0000293.s001
(PDF)
S2 Table. qPCR Data (Fig 2 and 3): Raw Cq values produced by the qPCR were used to calculate the DNA copies per microliter (Sq) using the standard curve equation shown below.
This equation was derived based on the Cq values for the postive controls.
https://doi.org/10.1371/journal.pwat.0000293.s002
(PDF)
S3 Table. Condensed Data (Fig 2 and 3): Mean copy number/µL and mean total copy numbers for each sampling site.
The total copy number was calculated by multiplying the mean copy number/µL by total sample volume which for FHS is 450µL and the commercial kit 100µL.
https://doi.org/10.1371/journal.pwat.0000293.s003
(PDF)
S4 Table. qPCR Data (Fig 4): qPCR was performed on both FHS and commercially extracted water samples over a period of five weeks to examine stability.
Raw Cq values produced by the qPCR method were used to calculate the DNA copy number/µL using the standard curve equation shown below. This equation was derived based on the Cq values for the postive controls.
https://doi.org/10.1371/journal.pwat.0000293.s004
(PDF)
S5 Table. Condensed Data (Fig 4): DNA stability across extraction methods over the course of five weeks.
Each data point represents six simultaneous water collections. Within each collection there are four technical replicates. A Student’s t-test was used to compare the SQ values each week, demonstrating no statistical significance, p > 0.05.
https://doi.org/10.1371/journal.pwat.0000293.s005
(PDF)
S6 Table. qPCR Data (Fig 5) Raw data collected using the FHS method with HF183 to detect human-associated FIB from Waller Creek in Austin, TX.
Copy number/µLwas converted to copy number/100mL for easier comparison with the FIB data.
https://doi.org/10.1371/journal.pwat.0000293.s006
(PDF)
S7 Table. FIB Data (Fig 5) E. coli were quantified using 3M Petrifilm.
Each sample site had two samples taken and each water sample was plated twice for a total of four readings per site.
https://doi.org/10.1371/journal.pwat.0000293.s007
(PDF)
S8 Table. Condensed Data (Fig 5) Values used to produce the graph.
https://doi.org/10.1371/journal.pwat.0000293.s008
(PDF)
S9 Table. Creek Mouth Distance (Fig 5) For each site along Waller Creek the distance from the mouth of the creek was measured in meters using Google Maps.
https://doi.org/10.1371/journal.pwat.0000293.s009
(PDF)
Acknowledgments
The authors would like to thank Ethan Glass, Christopher Pham, and Ammar Abed for their help in collecting data.
References
- 1. Sörme L, Lagerkvist R. Sources of heavy metals in urban wastewater in Stockholm. Sci Total Environ. 2002;298(1–3):131–45. pmid:12449334
- 2. Griffin DW, Donaldson KA, Paul JH, Rose JB. Pathogenic human viruses in coastal waters. Clin Microbiol Rev. 2003;16(1):129–43. pmid:12525429
- 3. Venkatramanan S, Chung SY, Kim TH, Prasanna MV, Hamm SY. Assessment and Distribution of Metals Contamination in Groundwater: a Case Study of Busan City, Korea. Water Qual Expo Health. 2014;7(2):219–25.
- 4. Olds HT, Corsi SR, Dila DK, Halmo KM, Bootsma MJ, McLellan SL. High levels of sewage contamination released from urban areas after storm events: A quantitative survey with sewage specific bacterial indicators. PLoS Med. 2018;15(7):e1002614. pmid:30040843
- 5. Mallin MA, Cahoon LB, Toothman BR, Parsons DC, McIver MR, Ortwine ML, et al. Impacts of a raw sewage spill on water and sediment quality in an urbanized estuary. Mar Pollut Bull. 2007;54(1):81–8. pmid:17064736
- 6. Soller JA, Schoen ME, Bartrand T, Ravenscroft JE, Ashbolt NJ. Estimated human health risks from exposure to recreational waters impacted by human and non-human sources of faecal contamination. Water Res. 2010;44(16):4674–91. pmid:20656314
- 7. Schoen ME, Soller JA, Ashbolt NJ. Evaluating the importance of faecal sources in human-impacted waters. Water Res. 2011;45(8):2670–80. pmid:21429551
- 8. Sidhu JPS, Ahmed W, Gernjak W, Aryal R, McCarthy D, Palmer A, et al. Sewage pollution in urban stormwater runoff as evident from the widespread presence of multiple microbial and chemical source tracking markers. Sci Total Environ. 2013;463–464:488–96. pmid:23831795
- 9. Ervin JS, Van De Werfhorst LC, Murray JLS, Holden PA. Microbial source tracking in a coastal California watershed reveals canines as controllable sources of fecal contamination. Environ Sci Technol. 2014;48(16):9043–52. pmid:25055204
- 10. Riedel TE, Thulsiraj V, Zimmer-Faust AG, Dagit R, Krug J, Hanley KT, et al. Long-term monitoring of molecular markers can distinguish different seasonal patterns of fecal indicating bacteria sources. Water Res. 2015;71:227–43. pmid:25618519
- 11. Paruch L, Paruch AM. An Overview of Microbial Source Tracking Using Host-Specific Genetic Markers to Identify Origins of Fecal Contamination in Different Water Environments. Water. 2022;14(11):1809.
- 12. Bernhard AE, Field KG. Identification of nonpoint sources of fecal pollution in coastal waters by using host-specific 16S ribosomal DNA genetic markers from fecal anaerobes. Appl Environ Microbiol. 2000;66(4):1587–94. pmid:10742246
- 13. Haugland RA, Varma M, Sivaganesan M, Kelty C, Peed L, Shanks OC. Evaluation of genetic markers from the 16S rRNA gene V2 region for use in quantitative detection of selected Bacteroidales species and human fecal waste by qPCR. Syst Appl Microbiol. 2010;33(6):348–57. pmid:20655680
- 14. Green HC, Haugland RA, Varma M, Millen HT, Borchardt MA, Field KG, et al. Improved HF183 quantitative real-time PCR assay for characterization of human fecal pollution in ambient surface water samples. Appl Environ Microbiol. 2014;80(10):3086–94. pmid:24610857
- 15.
U.S. EPA. Method 1696: Characterization of human fecal pollution in water by HF183/BacR287 TaqMan® Quantitative Polymerase Chain Reaction (qPCR) assay. 2019.
- 16. Riedel TE, Zimmer-Faust AG, Thulsiraj V, Madi T, Hanley KT, Ebentier DL, et al. Detection limits and cost comparisons of human- and gull-associated conventional and quantitative PCR assays in artificial and environmental waters. J Environ Manage. 2014;136:112–20. pmid:24583609
- 17. Kulski JK, Pryce T. Preparation of mycobacterial DNA from blood culture fluids by simple alkali wash and heat lysis method for PCR detection. J Clin Microbiol. 1996;34(8):1985–91. pmid:8818895
- 18. Myers DN. Distribution and variability of fecal-indicator bacteria in Scioto and Olentangy rivers in the Columbus, Ohio, area. U.S. Geological Survey.
- 19. Kim JY, Lee H, Lee JE, Chung M-S, Ko GP. Identification of human and animal fecal contamination after rainfall in the Han River, Korea. Microbes Environ. 2013;28(2):187–94. pmid:23666535
- 20. Deshpande JM, Shetty SJ, Siddiqui ZA. Environmental surveillance system to track wild poliovirus transmission. Appl Environ Microbiol. 2003;69(5):2919–27. pmid:12732567
- 21. Patel R, Babady E, Theel ES, Storch GA, Pinsky BA, St George K, et al. Report from the American Society for Microbiology COVID-19 International Summit, 23 March 2020: Value of Diagnostic Testing for SARS-CoV-2/COVID-19. mBio. 2020;11(2):e00722-20. pmid:32217609
- 22. Udugama B, Kadhiresan P, Kozlowski HN, Malekjahani A, Osborne M, Li VYC, et al. Diagnosing COVID-19: The Disease and Tools for Detection. ACS Nano. 2020;14(4):3822–35. pmid:32223179
- 23. Carducci A, Federigi I, Liu D, Thompson JR, Verani M. Making Waves: Coronavirus detection, presence and persistence in the water environment: State of the art and knowledge needs for public health. Water Res. 2020;179:115907. pmid:32389891
- 24. Larsen DA, Wigginton KR. Tracking COVID-19 with wastewater. Nat Biotechnol. 2020;38(10):1151–3. pmid:32958959
- 25.
Claman A, Jackson T, Richter A. Environmental integrity index phase I & II (2013-2014) watershed summary report. SR-15-08. City of Austin Watershed Protection Department. 2015.
- 26. Eichmiller JJ, Miller LM, Sorensen PW. Optimizing techniques to capture and extract environmental DNA for detection and quantification of fish. Mol Ecol Resour. 2016;16(1):56–68. pmid:25919417
- 27. Evans JS, López-Legentil S, Erwin PM. Comparing Two Common DNA Extraction Kits for the Characterization of Symbiotic Microbial Communities from Ascidian Tissue. Microbes Environ. 2018;33(4):435–9. pmid:30487350
- 28. Kinzelman J, Kay D, Pond K. Relating MST Results to Fecal Indicator Bacteria, Pathogens, and Standards. Microbial Source Tracking: Methods, Applications, and Case Studies. Springer New York. 2011. p. 337–59.
- 29. Korajkic A, McMinn BR, Harwood VJ. Relationships between Microbial Indicators and Pathogens in Recreational Water Settings. Int J Environ Res Public Health. 2018;15(12):2842. pmid:30551597
- 30. Lance RF, Guan X. Variation in inhibitor effects on qPCR assays and implications for eDNA surveys. Canadian Journal of Fisheries and Aquatic Sciences. 2018;77:23–33.
- 31. Kumar G, Eble JE, Gaither MR. A practical guide to sample preservation and pre-PCR processing of aquatic environmental DNA. Mol Ecol Resour. 2020;20(1):29–39. pmid:31633859
- 32.
City of Austin. Water quality sampling data | open data | city of Austin Texas; 2021 [cited 2021 Nov 18]. Database. Available from: https://data.austintexas.gov/Environment/Water-Quality-Sampling-Data/5tye-7ray
- 33. Sudarshan A, Rillera MG, Tran N, Riedel TE, Ellington AD, Bhadra S. Fieldable isothermal nucleic acid test for rapid semi-quantitative visual readout of Enterococci in recreational waters. BioRxiv. 2025.
- 34. Kildare BJ, Leutenegger CM, McSwain BS, Bambic DG, Rajal VB, Wuertz S. 16S rRNA-based assays for quantitative detection of universal, human-, cow-, and dog-specific fecal Bacteroidales: a Bayesian approach. Water Res. 2007;41(16):3701–15. pmid:17644149
- 35. Kapoor V, Lu T, Wendell D. Applying Molecular Tools for Microbial Source Tracking in the Duck Creek Watershed. proc water environ fed. 2012;2012(9):6541–61.
- 36. Mieszkin S, Furet J-P, Corthier G, Gourmelon M. Estimation of pig fecal contamination in a river catchment by real-time PCR using two pig-specific Bacteroidales 16S rRNA genetic markers. Appl Environ Microbiol. 2009;75(10):3045–54. pmid:19329663