| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Civil, Arch., and Environmental Engineering, Univ. of Miami, Miami, FL 33146
b Univ. of Florida, Gainesville, FL 32611
* Corresponding author (hmsolo{at}miami.edu).
Received for publication May 10, 2007.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
Directly monitoring for pathogens of concern, in addition to indicators, would provide the regulator with a more reliable basis to assess water quality at recreational beach areas. One of the main obstacles in pathogen monitoring is concentrating water samples for the pathogens of interest. Concentrating all three classes of pathogens (bacteria, protozoa, and viruses) simultaneously and efficiently in a standardized technique would be ideal to avoid time and cost delays as illness may result from any member of the three classes. The main approaches to simultaneously concentrate bacteria, virus, and protozoa are: size exclusion and membrane adsorption. Most waterborne pathogens range in size from 0.01 to 100 µm (Gerba, 1996). Methods based on size exclusion have been shown to be a promising option for concentrating all three classes of organisms (Paul et al., 1991, 1996; Hill et al., 2005; Olszewski et al., 2005). However, in many cases membranes clog prematurely during water filtration and elution of organisms from the membrane is sometimes difficult, especially at low concentration levels often observed in the environment. In addition, given the typical pore sizes of the membranes, e.g., 100 kDaA, they also concentrate a considerable amount of naturally occurring dissolved materials (e.g., humic and fulvic acids) and additional debris, which tend to cause inhibition in subsequent qPCR (quantitative polymerase chain reaction) analysis (Jiang et al., 2001).
Membrane adsorption has also shown promising results with respect to the concentration of viruses (Sobsey and Jones, 1979; Rose et al., 1984; Katayama et al., 2002; Scott et al., 2002; Fuhrman et al., 2005; Haramoto et al., 2005) and, potentially, bacteria and viruses simultaneously (Rolland and Block, 1980). In this method, viruses attach to charged membranes through electrostatic forces allowing for potential qPCR inhibitors to be removed through the filter permeate. The viruses can then be eluted off the membrane by altering the charge, resulting in cleaner samples, with less unwanted debris. However, the differences in the chemistry of the water filtered, e.g., pH and presence of cations, and the charge of the membrane cause significantly varied recoveries in adsorption to and elution off the membrane (Singh and Gerba, 1983; Sobsey and Glass, 1984; Sobsey and Hickey, 1985; Lukasik et al., 2000; Scott et al., 2002).
A system combining the advantages of both the size exclusion and membrane adsorption approaches is needed to simultaneously and efficiently concentrate different classes of pathogens and indicators. This system will also need to be relatively simple and easily integrated into commercial and regulatory routine analysis of indicator bacteria. Implementation of such a system would allow regulators to make better judgments concerning beach water quality and hence prevent more illnesses resulting from waterborne pathogens while minimizing economic losses associated with unnecessary beach closures.
The objective of this study was to evaluate the performance of filters when used in sequence for the concentration of a test bacterium and a test virus in marine waters. The vision that drives this evaluation is a bi-layer system in which two filters would be layered one over the other allowing for the simultaneous recovery of bacteria and viruses separately using the same water sample. The setup evaluated here is composed of two membranes in sequence, the first one designed to remove pathogenic and indicator bacteria based on size exclusion. This is referred to as the "top" membrane since it would be placed on top of the second membrane in the ultimate bi-layer design. The second filter is designed to capture viruses based on membrane adsorption. This filter is referred to as the "bottom" membrane since it would be on the bottom of the first filter in the ultimate bi-layer design. Experiments were conducted to determine the capability of different membranes in their ability to either adsorb or allow passage of a test virus and test bacteria. The effects of filter volume and the effects of pre-conditioning the membranes with beef extract solution were assessed. Elution of the test virus was evaluated through two commonly used elution techniques, one based on a combination of acid and base and the other based on the use of beef extract (USEPA, 2001; Katayama et al., 2002; APHA, 2005). The results of this study are intended to serve as a proof-of-concept, recognizing that additional experimentation is necessary before such a system can be fully implemented.
|
Materials and Methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
Initial Sample Preparation
Experimentation occurred over the course of six different dates covering a 10-mo time period. During the morning of each experiment, 1 to 10 L of beach water (
4.5 NTU and a salinity of 36 
on average) were collected from knee-depth water at Hobie Cat Beach and 1 to 2 L of untreated sewage were collected from a local wastewater treatment facility (Miami-Dade Central District Wastewater Treatment Plant). The sewage was filtered through a 90-mm diameter 25 µm pore size glass fiber filter (VWR, Grade 415 filter paper) to remove larger suspended solids. This coarsely filtered solution was used to spike the beach water. "Unspiked samples" were used for experiments focusing on the performance of the top filter for enterococci analysis. "Spiked samples" consisted of ocean water spiked with 10% of the raw coarsely filtered sewage to ensure the presence of somatic coliphage. The spiked water sample was vigorously mixed and subsamples were used in the filtration experiments for coliphage. All filtrations in this study were conducted at a rate of 1.7 to 3.3 mL/sec.
Membranes
Four types of 0.45 µm pore size membranes were used in this study. All were 47 mm in diameter. For the top membrane a polyvinylidene fluoride (PVDF) membrane (Millipore, Durapore-hydrophilic) characterized by low protein binding was used. Because of these properties, the PVDF membrane was hypothesized to permit the passage of coliphage but retain the enterococci. A high protein binding type HA membrane with a mixed cellulose ester composition (Millipore, MF-Millipore) was used as the bottom membrane. Given this membrane's properties it was hypothesized that it would adsorb the coliphage which permeates through the top membrane (Lukasik et al., 2000; Katayama et al., 2002; APHA, 2005). Two other membranes (Pall, GN-6 Metricel and Whatman, Cellulose-nitrate), traditionally used for the enumeration of enterococci in water, were evaluated for comparative purposes. Both of these membranes are made of cellulose nitrate/acetate mixtures. Given their properties it was suspected that they would adsorb coliphage. The Pall membrane was compared to the PVDF membrane for enterococci enumeration and both the Pall and Whatman membranes were compared to the HA membrane for coliphage adsorption. The maximum volume of spiked samples that could be filtered through the membranes without significant clogging was between 300 and 400 mL.
Evaluation of Top Membrane for Enterococci Enumeration
The performance of the PVDF membranes was compared to the performance of the traditional Pall membrane for their capacity to provide similar enterococci counts. To accomplish this comparison, unspiked samples were split into 10- to 100-mL aliquots. These aliquots were then filtered through each type of filter in triplicate and analyzed according to the standard USEPA Method 1600 (USEPA, 2002). In brief, analysis for enterococci included the filtration of the predetermined volume of sample followed by a 20 to 40 mL rinse with phosphate buffered saline solution. The membranes were then placed right side up on Petri plates containing mEI agar, incubated for 24 h at 41 ± 0.5°C, and blue colonies and colonies with blue halos were counted and reported as colony forming units (CFU) per 100 mL.
Evaluation of Top Membrane for Passage of Coliphage
Evaluation of the top membrane for the passage of coliphage included the measurement of coliphage retained on the membranes and the measurement of coliphage observed within the permeate. A 47 mm magnetic filter holder attached to a 1 L side arm flask and vacuum pump was used as the filtration apparatus. Two sets of experiments were conducted. The first evaluated the need for pre-conditioning the top membranes; the second focused on evaluating traditional membranes (used for enumerating bacteria) for the passage of coliphage.
To evaluate the need for pre-conditioning the membranes, four filter holders were set up with PVDF membranes with two of the setups preconditioned. Conditioning the membranes consisted of passing 10 mL of autoclave-sterilized 1.5% beef extract solution (1.5 g beef extract powder and 0.375 g glycine in 100 mL of deionized water adjusted to a pH of 7.3 ± 0.2) as described in APHA standardized procedures method 9224 F (APHA, 2005). The permeate from the beef extract conditioning was collected and retained in the receiving side-arm flask as conditioning of the filter would affect subsequent sequential filtration within a dual membrane system. One hundred mL of spiked sample were filtered through one of the membranes conditioned with beef extract solution and one without conditioning. Four hundred milliliters of spiked sample were filtered through each of the remaining two membranes (Fig. 1 ). The four membranes and the permeates from these membranes were stored separately at 4°C for up to 5 h until further use and/or analysis. These four filtration setups were repeated two times for a total of three complete experiments for each setup.
|
Adsorption of Coliphage on Bottom Membrane
Evaluation of the bottom membrane for the adsorption of coliphage included the measurement of coliphage retained on the membranes and the measurement of coliphage observed within the permeate. Identical filter setups were used for the bottom membrane as used for the top membrane except that these sets of four filter holders were fitted with HA membranes. In this case, 100 and 300 mL of spiked sample as well as permeates from the PVDF membrane were passed through each of the HA membranes (Fig. 1). The permeates from the direct application of samples through the HA filters and re-filtered permeates (first passing the PVDF and then the HA membrane) were stored at 4°C for up to 5 h until analysis. These four filtration setups were repeated two more times for a total of three complete experiments for each setup.
Elution of Coliphage from Bottom Membrane
Three elution experiments were conducted using the bottom membrane. The first experiment involved filtering 100-mL aliquots of spiked sample (not pre-filtered through a PVDF membrane) through four filter holders fitted with HA membranes. Two of the four membranes were then rinsed with 10 mL of 0.5 mmol L–1 H2S04 (pH = 2.3 to 2.7) and the permeate was discarded. The purpose of the acid rinse was to remove cations, which bind the viruses to the membrane and allow viruses to attach directly to the membrane (Katayama et al., 2002). All four holders were then placed on smaller receiving reservoirs (25 mL side-arm flasks) and eluted with 5 mL of either 1 mmol L–1 NaOH, (pH = 10) or 1.5% beef extract solution (pH of 7.5) to release the viruses from the membrane and into the elution solution. For each elution solution, two membranes were eluted, one with and one without the prior acid rinse. The NaOH eluate was neutralized on elution with 25 µL of 100x Tris-EDTA (TE) buffer (pH = 7.4 to 7.7) and 25 µL of 50 mmol L–1 H2S04 (pH = 1.6). Eluates as well as the membranes were stored at 4°C for up to 5 h until further use and/or analysis.
The second experiment also involved filtering 100-mL aliquots of spiked sample through four filter holders fitted with HA membranes. However, because recovery was less than expected in the first experiment, several parameters were modified when formulating the second experiment: elution solution volume was increased from 5 to 10 mL, elution rate was changed from a quick filtration (1.7 mL/sec) to a drop by drop method where most of the elution solution was allowed to slowly permeate through the membrane without a vacuum to increase contact time, the spiked sample was pre-filtered through an unconditioned PVDF membrane before filtration through the HA membrane, and the pH of the beef extract elution solution was raised from 7.5 to 9.
The objective of the third experiment was to determine if loss of recovery during coliphage elution in the first two experiments resulted from lack of elution from the membrane or deactivation of the coliphage by the elution solutions. A simple experiment was set up to determine the cause of this loss using a mass balance approach by quantifying the amount of coliphage entering the system, the amount attached to the HA filter, the amount in the permeate before elution, and the amount in the eluate. These experiments involved setting up four filter holders with HA membranes as described above. Spiked samples were diluted 1:50 using sterile (autoclaved) beach water, and 10 mL were pre-filtered through unconditioned PVDF membranes. This decrease in sample volume and dilution was required to obtain countable plaques on the membranes. The permeates from the PVDF membranes were subsequently filtered through each of the HA membranes. The elution method was the same as that used for experiment 2 described in the previous paragraph. The PVDF permeate (input), HA permeate before elution, HA membrane, and HA eluate were analyzed for coliphage.
Analysis of Somatic Coliphage by the Single Agar Layer Method
Permeate and eluate, as well as the direct original spiked sample, were analyzed for somatic coliphage using standardized method 9211 D (APHA, 1998). In brief, Escherichia coli C (ATCC#13706), the host culture, was propagated at 35°C on tryptic(ase) soy agar (pH of 7.3 ± 0.1 at 25°C). The host was then inoculated into tryptic(ase) soy broth at a pH of 7.3 ± 0.1 at 25°C, and incubated at 35°C until an optical density of 0.5 at 520 nm was obtained as measured using a spectrophotometer (Milton Roy Co., Spectronic 20). The host was not reinoculated at this point but rather stored at –20°C for up to a maximum of 12 wk. A decrease in the viability of the E. coli was not observed during this period of time, since the bacterial lawn was evident. Five and a half milliliters of modified tryptic(ase) soy agar were then mixed with 1 mL of thawed host medium and 0.08 mL of a 1% 2,3,5-triphenyl tetrazolium chloride (TPTZ) solution (Sigma-Aldrich, St. Louis, MO) to enhance plaque visibility. Five mL of sample were added to this mixture and then poured into 100 mm Petri dishes. Samples suspected to contain high coliphage counts were also diluted 1:10 or 1:100 with phosphate buffer saline pH of 7 ± 0.3 before addition to the agar mixture. Plates were inverted and incubated at 35°C for 24 h, as opposed to the 4 to 6 h incubation recommended in the method to allow for further development of plaques. Plaques were counted and reported as plaque forming units (PFU) per 100 mL of sample.
Analysis of Somatic Coliphage by the Membrane Filtration Technique
A standardized membrane filtration method 9224 F was used to confirm the adsorption, passage, or elution of coliphage on both the top and/or bottom membranes by directly analyzing the membrane (APHA, 2005). In this technique, membranes from spiked sample filtrations were placed face down on modified tryptic(ase) soy agar plates containing 2 mL of E. coli C, 1 mL of Tween 80, and 1 mL of TPTZ per 100 mL of agar media. TPTZ was used instead of the recommended tetrazolium violet, which is also used to enhance visibility. It was found that the TPTZ provided clearer plaques as compared with the tetrazolium violet. Membranes with media were then incubated at 36.5 + 2°C overnight and plaques were then counted and reported as PFU/100 mL.
|
Results |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
Evaluation of Top Membrane for Enterococci Enumeration
There was no statistical (p > 0.05) difference between the PVDF and Pall membranes in detecting colony counts (CFU) for enterococci. For the 100 mL filtration of unspiked beach water the PVDF membrane recovered a mean of 11 ± 6 CFU/100 mL, while the Pall membrane recovered 19 ± 15 CFU/100 mL. Enterococci colonies forming on the PVDF membrane appeared slightly smaller and more faint in color than on the conventional Pall membrane mainly because of the fact that the PVDF membrane is clear, and not opaque white as the Pall membrane.
Evaluation of Top Membrane for Passage of Coliphage
For the PVDF membrane, over 99% of the somatic coliphage passed through the membrane as observed from the analysis of the permeate. Increasing the volume from 100 to 400 mL or conditioning the membrane with beef extract solution did not have a significant (p > 0.05) effect as greater than 99% on average of somatic coliphage passed in all cases. Thus beef extract solution conditioning was not required for the top PVDF membrane (Table 1
).
|
Adsorption of Coliphage on Bottom Membrane
The HA membrane adsorbed 89 and 91% of somatic coliphage on average for the 100 and 300 mL filtration of direct unspiked sample, respectively, suggesting that typical ranges in sample volumes for 47 mm diameter filters did not have a significant (p > 0.05) effect on adsorption efficiencies. Seventy eight percent of the somatic coliphage was adsorbed onto the HA membrane when pre-filtered through the PVDF membrane. Comparing the results in Table 1 shows that the presence of beef extract conditioning (top membrane) in the sample resulted in a significant (p < 0.05) decrease of 72%, (94–22%), in coliphage adsorption on the HA membrane. The use of beef extract should thus be avoided when subsequent adsorption is desired on the bottom membrane. The use of PVDF filters as the top membrane eliminates the need for the use of beef extract conditioning as these membranes allow for over 99% somatic coliphage passage without conditioning as indicated above.
Observing Somatic Coliphage Attached to the Membrane
In the initial experiments, both top and bottom membranes were characterized by overlapping plaques when membranes were placed directly on the host lawn due to the large number of somatic coliphage captured. However, in a subsequent experiment, conducted as part of the third elution experiment, a smaller filtration volume of a diluted spiked sample resulted in quantifiable plaque counts on the top filter. The numbers observed on the top PVDF membrane corresponded to 3% of the somatic coliphage, while the HA membrane showed overlapping plaques which were too numerous to count, indicating that a much larger proportion were captured by the bottom membrane.
Elution of Coliphage from Bottom Membrane
From the first two elution experiments (Table 2
), beef extract without an acid rinse was observed to yield the highest recovery (35%), while the lowest recovery (2%) resulted from elution with NaOH with an acid rinse. The third mass balance experiment (Table 3
), which was focused on evaluating the reason for this recovery loss using diluted, small volume sample filtrations, resulted in much higher recoveries in the eluate of the samples. The highest recovery (84%) resulted from a NaOH elution without an acid rinse. It was evident that substantial loss of somatic coliphage, up to 71%, occurred in the presence of the acid rinse. This loss is presumably due to the inactivation of the coliphage by the acid rinse. However, increasing filtration volumes from 10 to 100 mL shows a significant loss of recovery even in the absence of acid. This is evident in the fact that beef extract and NaOH, without an acid rinse, yield recoveries of 35 and 11% (Table 2) in the 100 mL filtrations and 55 and 84% in the 10 mL filtrations, respectively (Table 3).
|
|
|
Conclusions |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
The method described in this paper was developed using somatic coliphage from environmental samples as the test non-pathogenic organism to determine virus adsorption recoveries and enterococci as the test culturable bacteria to assess the ability of the membrane to filter bacteria. Both of these organisms were intended to serve as proof of concept and actual pathogens of interest should be used to test recovery efficiencies on a site-specific basis before implementation of this method.
Beach water samples with and without a sewage spike (as opposed to direct organism spikes) were used to simulate a scenario where water would be contaminated with sewage. The configuration of this setup consisted of a top PVDF membrane which was intended to filter out bacteria while allowing viruses to pass without attachment. Analysis of the top membrane's permeate showed that over 99% of the somatic coliphage passed through this low protein binding membrane. This high recovery may be explained by the fact that there are fewer microbes in the permeate of the PVDF membrane allowing for less competition and interference and hence more effective plaque formation than when plating the direct sample. Preliminary research with autoclaved sterile ocean water spiked with MS2 coliphage shows a similar phenomena with the permeate of the PVDF resulting in higher counts than the input (Abdelzaher, unpublished observation). Therefore, the PVDF appears to remove microbes and other particles larger than 0.45 µm, which may interfere with plaque formation. This can be viewed as an advantage of the PVDF membrane since it allows for improved virus detection but could also be considered a limitation since input counts of the somatic coliphage may be underestimated. This finding will require further investigation to evaluate why recoveries increase after filtration through the PVDF membrane.
The bacteria filtration step was successful as enterococci counts were statistically equivalent in both the proposed PVDF membrane and a conventional membrane. However, experiments should be conducted with other types of bacteria to determine if the PVDF membrane causes any inhibitory effects on their growth. The bottom HA membrane was intended to capture viruses from the permeate of the top membrane. On average, 78% of the somatic coliphage from the top PVDF membrane's permeate attached to this HA membrane. Somatic coliphage was also measured directly on the membranes and confirmed that almost all the somatic coliphage passed through the PVDF membrane and the majority attached to the bottom HA membrane. For both the top and bottom membranes, varying the volume filtered from 100 to 400 mL did not have a significant effect on the somatic coliphage recovery. Conditioning the top membrane with beef extract solution did not significantly increase the amount of coliphage that passed through the top membrane, but significantly decreased the amount of coliphage that adsorbed onto the bottom membrane and therefore should be avoided in this setup. In addition, beef extract may also inhibit PCR (De Leon et al., 1990, Schwab et al., 1991) as reviewed by Tsai et al. (1993). So eliminating the use of beef extract has the added advantage of minimizing interferences if PCR is the chosen detection method.
Elution of viruses from the bottom membrane resulted in the largest loss of the coliphage recovery. When filtering 100 mL, a maximum of 35 and 16% of the attached coliphage were eluted off the membrane with beef extract solution and NaOH, respectively, without an acid rinse. Elution recovery results were lower in these experiments when compared to other findings using either the acid/base elution or a beef extract elution (Katayama et al., 2002; Scott et al., 2002). In the case of the acid/base elution the decrease in recovery may be due to the fact that coliphage is being used as the test organism instead of more acid-resistant viruses such as poliovirus (Katayama et al., 2002). Further investigation is needed to determine the ideal elution method depending on the specific viruses which are concentrated. Conducting elution experiments with smaller sample volumes (10 ml) showed that elution was more effective with up to 84% of the somatic coliphage eluted with an NaOH solution alone. The inclusion of an acid rinse step during elution was found to inactivate up to 71% of the somatic coliphage. This observation is consistent with Sabatino and Maier (1980) who observed deactivation of somatic coliphage in the presence of acids. Therefore, both increasing the volume of the sample filtered and rinsing with H2S04 before elution significantly decreased the elution recovery of viable somatic coliphage. Inactivated phage, however, may still contain intact nucleic acids which may have been eluted. This is an important fact since inactivation is not a deterrent in qPCR analysis and therefore if qPCR is used to analyze the eluent, acid may not be a source of recovery loss.
The advantages of the proposed method include the ability to sequentially concentrate two different classes of microbes (bacteria and virus) separately using the same water sample to provide for the removal of large particles (>0.45 um) in the top filter. The bottom filter is thus comparatively clean of large debris and contains the viruses, thus decreasing inhibitory effects during subsequent virus analysis. Subsequent research will focus on designing a bi-layer filter holder, which would allow for the placement of both the top and bottom membranes in series so that filtration of the water would concentrate both the bacteria and viruses in one step (Fig. 2 ).
|
This research demonstrates a step toward analyzing water for multiple microbial groups directly and simultaneously to expand analyses beyond measurements of bacterial indicators alone. Increasing the suite of microbes evaluated will better define microbial populations in the water and provide useful information that will allow regulators to make better informed decisions when evaluating beach closures. Future experiments should further validate the method through direct detection of pathogens and the evaluation of water samples with different chemical, physical, and biological properties as well as using larger size (90 mm) membranes which would allow for higher and more realistic filtration volumes. Specifically, the effect of suspended solids on the recoveries should also be assessed. The beach water utilized in this study was characterized by relatively low suspended solids (4.5NTU). Of interest would be to evaluate more turbid waters using the sequential filtration setup.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. M. Abdelzaher, M. E. Wright, C. Ortega, H. M. Solo-Gabriele, G. Miller, S. Elmir, X. Newman, P. Shih, J. A. Bonilla, T. D. Bonilla, et al. Presence of Pathogens and Indicator Microbes at a Non-Point Source Subtropical Recreational Marine Beach Appl. Envir. Microbiol., February 1, 2010; 76(3): 724 - 732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Abdelzaher, H. M. Solo-Gabriele, C. J. Palmer, and T. M. Scott Simultaneous Concentration of Enterococci and Coliphage from Marine Waters using a Dual Layer Filtration System J. Environ. Qual., October 29, 2009; 38(6): 2468 - 2473. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
.