WO2002099128A2

WO2002099128A2 - Sequence independent identification of viruses

Info

Publication number: WO2002099128A2
Application number: PCT/GB2002/002606
Authority: WO
Inventors: David John Griffiths; Paul Kellam; Robert Anthony Weiss; Peter Simmonds
Original assignee: University College London
Priority date: 2001-06-07
Filing date: 2002-06-07
Publication date: 2002-12-12
Also published as: WO2002099128A3; AU2002311437A1; GB0113909D0

Abstract

A method for identifying a virus in a biological sample from a human is provided, the method comprising subjecting the sample to conditions which purify viral nucleic acid, shotgun cloning the purified viral nucleic acid, sequencing the cloned nucleic acid, and determining whether the sequence of the cloned nucleic acid is present in, or absent from, at least one human genomic database; wherein presence of the sequence in the database indicates that the cloned nucleic acid is a human nucleic acid which contaminated the purified viral nucleic acid, and absence of the sequence from the database indicates that the cloned nucleic acid may be a viral nucleic acid. If the sequence is absent from the database, the nucleic acid may then be studied further to confirm or deny that it is form an unknown virus.

Description

SEOUENCE INDEPENDENT IDENTIFICATION OF VIRUSES

The invention concerns the identification of viruses in human biological samples using recombinant DNA techniques.

Background to the invention

Biological materials can often become contaminated with unidentified viruses. For example, cells grown in tissue culture often exhibit signs of a cytopathic effect consistent with a virus infection but the identity of the virus may not be apparent. Human blood products, such as factor VIII for the treatment of haemophiliacs, can be contaminated with unidentified viruses, as was demonstrated by infection of many haemophiliacs with human immunodeficiency virus in the early 1980s. Similarly, two decades ago 20% of individuals who received transfused blood contracted hepatitis C (Randall, 2001, J Pediatr. Oncol. Nurs. 18(1), 4-15).

Summary of the invention

The invention provides a method for screening a biological sample from a human for the presence of a virus, comprising subjecting the sample to conditions which purify viral nucleic acid, shotgun cloning the purified viral nucleic acid, sequencing the cloned nucleic acid, and determining whether the sequence of the cloned nucleic acid is present in, or absent from, at least one human genomic database; wherein

- presence of the sequence in the database indicates that the cloned nucleic acid is a human nucleic acid which contaminated the purified viral nucleic acid, and

- absence of the sequence from the database indicates that the cloned nucleic acid may be a viral nucleic acid. If the sequence of the cloned nucleic acid is absent from the human genomic database the nucleic acid may then be subjected to further investigation to confirm or deny that it is from a virus. Such further investigation may involve determining whether the sequence of the nucleic acid is present in, or absent from, at least one database comprising sequences of viruses. If the sequence is absent from both the human genomic database and the viral database, this suggests that the sequence may be from a previously unknown virus.

The method of the invention may include a further step of experimentally determining whether the sequence of the nucleic acid is present in human DNA by making a probe comprising at least part of the sequence of the nucleic acid and determining whether the probe hybridizes to human DNA, and/or making a set of primers comprising sequence from the nucleic acid and determining whether the primers are able to amplify human DNA. Furthermore, DNA walking may be carried out to determine any sequence which flanks the sequence of the cloned nucleic acid. If it is confirmed that a virus identified by the method of the invention was previously unknown, then assays can be performed to determine whether the virus is associated with a disease.

Detailed description of the invention

There are a number of human diseases in which unidentified viruses are thought to play a causative role. For example, unidentified viruses are believed to play a role in cancers such as leukaemia, autoimmune diseases such as rheumatic disease, cardiovascular diseases such as dilated cardiomyopathy and Kawasaki disease, and prostatitis (zur Hausen 2001 The Lancet 357, 381-384; Greaves 1997 The Lancet 349, 344-349; Rowley and Shulman 1998 Clinical Microbiology Reviews 11(3), 405-414; Kawai 1999 Circulation 99, 1091-1100; and Dominigue and Hellstrom 1998 Clinical Microbiology Reviews 11(4), 604-613). Particles- resembling retroviruses have been reported in affected tissue from patients with psoriasis, Sjδgren's syndrome and rheumatoid arthritis (Iverseh 1990 J. Invest. Dermatol. 90, 41S-3S; Garry et al 1990 Science 250, 1127-9; Yamano et al 1997 J. Clin. Pathol. 50, 223-30; and Stransky et al 1993 Br. J. Rheumatol. 32, 1044-8). The invention provides a way of screening for the viruses which may cause or contribute to such diseases. Once identified, the viruses may be used as a target for developing diagnostic tests for, or therapies against, the diseases.

The first step in the method of the invention is to purify viral nucleic acid from the biological sample. The biological sample may be any human sample which is susceptible to infection by a virus. The sample may be any of a range of tissue and fluid types, for example blood serum, seminal fluid, breast milk, saliva, cerebrospinal fluid, urine, bile, bronchial lavage fluid, nasal secretion, eye secretion or vaginal wash. The sample may be derived from one individual or may be a pool of material from more than one individual (up to e.g. 10, 100, 1000 or 10,000 individuals). The use of pooled material may be advantageous since it allows material from a large number of individuals to be tested at once, thereby reducing the number of experiments which need to be performed.

Generally, whole genomic nucleic acid of the virus is purified in the first step. The purification step does not need to lead to a total purification of the nucleic acid; on the contrary the purification may be a relatively crude process. It is enough that the purification enriches the content of the viral nucleic acid to an extent that it is no longer overwhelmed by other nucleic present in the biological sample. The purification should be such that it enriches the content of viral nucleic acid relative to human nucleic acid; this means that, when the shotgun cloning is carried out, the viral nucleic acid is not overwhelmed by human nucleic acid.

The purification of viral nucleic acid may include a step of treating a suspension comprising the virus with a nuclease so as to digest extraneous nucleic acid, wherein the viral nucleic acid is protected from digestion by viral coat or core protein. The nuclease is preferably a non sequence-specific nuclease which digests DNA and/or RNA, for example micrococcal nuclease S7 (Roche Molecular Biochemicals, Catalogue 107 921).

Prior to treatment with the nuclease, the viral nucleic acid may be concentrated by centrifuging the biological sample under conditions such that cell debris is pelleted and virus particles remain in the supernatant; collecting the supernatant; and centrifuging the supernatant under conditions such that virus particles are pelleted.

The initial centrifugation to pellet the cell debris may, for example, be carried out at 100 to 10,000 g, preferably from 1000 to 10,000 g. The subsequent centrifugation to pellet the virus particles is carried out at a higher g force, for example 50,000 to 500,000 g, preferably about 100,000 g.

Once the viral nucleic acid has been purified, the next step is to subject it to shotgun cloning. In essence, this involves inserting random sections of the viral nucleic acid into a vector in which the sections can be sequenced. The shotgun cloning may be accomplished in a number of ways. For example, the nucleic acid may be fragmented with a restriction enzyme and the fragments then inserted into a vector. However, the cloning is generally facilitated by an amplification step, such as a PCR step, which amplifies random sections of the nucleic acid. The amplified sections may then be cloned into the vector.

Random PCR may be directed using a primer which directs initiation of DNA synthesis at random sequences. Such a primer may be made by synthesising it so that it contains a random sequence, for example a sequence of at least 6 consecutive nucleotides (e.g. from 6 to 20 nucleotides) wherein each nucleot-ide may be any of the four natural nucleotides, i.e. A, T, C or G. In other words, such a primer contains a sequence NNNNNN wherein each N is A, T, C or G.

In some cases the target nucleic acid is RNA, for example in cases where the target virus is an RNA virus. In such cases, it is necessary to reverse transcribe the RNA to produce a complementary DNA strand and then synthesise a second DNA strand before carrying out PCR. This can be achieved by using a primer which directs initiation at random sequences in a reverse transcription reaction and then a second strand synthesis reaction. Such a primer may be synthesised using a similar principle to that described above for random PCR primers. The reverse transcription primer and the second strand synthesis primer generally include a specific known sequence of nucleotides in addition to the random sequence, so that the known sequence becomes incorporated into each end of the double-stranded DNA and can then be used as the target for PCR primers to amplify the double-stranded DNA.

The vector into which the random sections of nucleic acid are cloned may, for example, be a plasmid or a bacteriophage vector. Suitable plasmids are known and commercially available, such as pBluescript™ (Stratagene) and pGEM-T-Easy™ (Promega). Suitable bacteriophage include bacteriophage λ and Ml 3.

After sequencing the cloned sections of nucleic acid, the next step is to determine whether each sequence is present in at least one human genomic database. At least one of the databases searched is generally a comprehensive or consensus human genome database. Preferably, at least one of the databases contains an essentially complete human genome sequence. It is necessary to do this because the shotgun cloning procedure results in pieces of human DNA being cloned which may not be of interest and need to be discarded from further consideration. However, it needs to be borne in mind that, although there has recently been a great deal of publicity about the "completion" of the human genome sequence, not all the human genome has in fact been sequenced, and it is possible that a cloned sequence could fall within the unsequenced part of the genome. The human genome contains large areas with repetitive sequences, and much of the unsequenced genome is within these areas. In order to make as comprehensive a search as possible, it is desirable to search a range of different types of database; in addition to a human genome database, it is desirable to search, for example, a database comprising expressed sequence tags (ESTs) and a database comprising repetitive elements of the human genome. Appropriate databases include GenBank, the EMBL database, the Celera human genome database, the Ensemble human genome database, the DNA Data Bank of Japan (DDBJ), the Incyte LifeSeq™ database of ESTs and the Repbase database of repetitive elements in the human genome.

The databases will also generally include a database comprising sequences of viruses. Appropriate databases include the virus subdivision of GenBank and the VIDA database (Alba et al 2001 Nucleic Acids Research 29(1), 133-136).

Where the cloned nucleic acid sequence is found to be not present in any of the interrogated databases of known sequences, this indicates that the nucleic acid may be from a previously unknown virus. The nucleic acid then becomes a candidate for further investigation and may be designated a Primary Candidate Virus (PCV).

Although the nucleic acid sequences designated PCVs have failed to match any sequence in databases of known sequences, this does not necessarily mean that a given PCV is from a previously unknown organism because not all the sequences of all known organisms are present in databases. For example, as montioned above, not all the human genome has in fact yet been sequenced, and it is possible that a nucleic acid designated a PCV may fall into an unsequenced part of the human genome.

It is therefore generally necessary to confirm by experimentation that a nucleic acid sequence designated a PCV is not in fact a human sequence. A preferred way of doing this involves designing and synthesising a specific primer- set (or sets) to amplify the nucleic acid designated a PCV and determining whether the set(s) are able to amplify any DNA in a sample of complete genomic human DNA. The amplification conditions for each primer set may be optimised using the original cloned nucleic acid fragment, for example by testing the primer set under a number of different buffer conditions and at a number of different temperatures (typically from 45 °C to 65 °C). The PCR system may be used to screen one or more samples of human genomic DNA, for example from 1 to 100 samples, preferably from 5 to 50 samples. As an alternative to PCR, human genomic DNA may be probed with a labelled probe containing sequence from the original cloned nucleic acid (e.g by Southern blotting).

If the PCV cannot be detected in human DNA by experimentation (by PCR or hybridisation with a labelled probe), it may then be subjected to further analysis. It may be designated a Secondary Candidate Virus (SCV).

The further analysis of an SCV may include DNA walking to determine whether the original cloned nucleic acid sequence exists in nature as part of a longer sequence, such as the genomic sequence of an unknown virus. DNA walking may be carried out using techniques known in the art, such as vectorette PCR (Allen et al, PCR Methods Appl. 4:71-75), rapid amplification of cDNA ends (RACE, Frohman et al Proc Natl Acad Sci U S A..85:8998-9002), rapid amplification of genomic ends (RAGE, Cormack and Somssich. 1997. Gene. 194:273-276) and methods derived from these. Alternatively, the SCV sequence may be "extended" by screening a DNA library using the original cloned nucleic acid sequence as a probe.

The additional sequence information obtained through DNA walking may reveal information about the identity of the SCV which cannot be determined from the original clone. The additional information may therefore be analysed, for example to determine whether it contains an open reading frame (i.e. a sequence encoding a protein); the presence of an open reading frame provides further support for the suggestion that the SCV is a virus. Furthermore, the additional information may identify the SCV as being related to a known virus; for example, the information may identify the SCV as being a new member of a known family of viruses.

A further step may then be to determine whether a newly-identified candidate virus is associated with a disease, for example with a cancer, autoimmune disease, cardiovascular disease or other disease mentioned above. This may be done by obtaining a specimen from each member of a group of subjects with a disease; determining whether the cloned nucleic acid or other nucleic acid of the same virus is present in each specimen; and determining whether the proportion of subjects in whom the nucleic acid is present is greater in the group of subjects who have the disease than in a control group of subjects who do not have the disease, wherein a said greater proportion suggests that the virus may cause or contribute to the disease.

Typically, the process of determining whether the nucleic acid is present or absent from a specimen from a subject may be carried out by PCR using primers specific for the novel sequence (including any contiguous sequence obtained by DNA walking). Initially, perhaps from 10 to 50 patients from a disease group may be tested, but if positive results are obtained in initial studies, the investigation may be extended to a larger group (e.g. a group of up to 100, 500, 1000 or 10,000). The nature of the biological specimens taken from the members of the group varies depending on the disease association that is being investigated; where possible specimens are from disease affected tissue and from peripheral blood of the subjects (for a published example of this see Griffiths et al, 1999, Arthritis Rheumatism, 42:448-454). The specimens may be from the same tissue and fluid types as the biological samples used in the initial screening assay described above.

Once a new virus has been identified and found to be positively associated with a particular disease or condition, serological and genetically-based diagnostic assays for infection by the virus may readily be devised. Genetically-based assays can be developed by using the nucleotide sequence of the virus to design probes and/or PCR primers for specifically detecting the nucleic acid of the virus. Serological assays can be developed by producing recombinant proteins or protein fragments encoded by the virus and testing for the presence of antibodies to these proteins in human sera. Alternatively, antibodies specific for the proteins of the virus may be made and the antibodies used to, detect the virus directly. The serological assays may take the format of an ELISA, western blot or immunofluorescence assay. Correlations may be sought between serological data and genetic data. Furthermore, the virus provides a target for the development of therapies and/or prophylactic vaccines against the disease.

Examples

Example 1 : Identification of an unknown virus in tissue culture supernatant

The application of the cloning strategy of the invention has been demonstrated in a tissue culture system. A culture of U87 cells (a human glioma cell line) was noted to (apparently spontaneously) exhibit signs of a cytopathic effect consistent with virus infection. Transfer of the cytopathic effect to uninfected A549 cells (a human lung epithelial cell line) by filtered culture medium confirmed the presence of a virus in this culture. Initial PCR reactions with primers specific for viruses commonly used in the laboratory of the Inventors failed to identify the suspected viral agent. In order to identify the virus, the following method was applied.

The A549 culture supernatant was filtered through a 0.2 μm filter and a virus pellet was prepared from 500 μl of the filtrate by centrifugation first at low speed (4000g, 10 minutes 4°C) to remove cell debris and then at high speed to pellet viral particles (100,000g, 15 minutes, 4°C),

Due to the high level of cell death in the culture it is likely that some free nucleic acids were present in the culture medium. To remove this extraneous DNA and RNA, the pellet was gently resuspended in 100 μl miclease buffer (3 mM Tris-Cl pH 8.8, 2mM CaCl₂) and 75 units of micrococcal nuclease S7™ (Roche Molecular Biochemicals, Catalogue 107 921) were added and the reaction was incubated for 30 minutes at 37°C. Any nucleic acids remaining in the pellet were then purified. Half of the reaction was treated for DNA purification by proteinase K digestion and phenol chloroform extraction using a standard method (Sambrook et al., (1989) Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, New York). The other half was treated for RNA purification by extraction in acidified chloroform (Chomsznski, P. and Sacchi, N. (1987) Anal. Biochem,. 162, 156-159) using the RNAZol B™ reagent (Biotecx Laboratories Inc.) essentially as recommended by the manufacturers.

The unknown virus present in the original culture was subsequently identified by analysis of the RNA fraction. To rapidly facilitate cloning of the RNA sequences a sequence independent PCR strategy was employed. This method is essentially as described by Froussard (Froussard, P. (1993). rPCR: a powerful tool for random amplification of whole RNA sequences. PCR Methods Appl, 2, 185-190) and provides a more convenient method for cloning small amounts of RNA than do conventional methods for constructing cDNA libraries (see e.g., Sambrook et al., 1989). Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, New York). 10 μl of purified RNA (50% of the total) was mixed with 100 ng of the oligonucleotide primer UP-dN6 (5'-

GCCGGAGCTCTGCAGAATTCNNNNNN, where N is an A, a C, a G or a T) and heated to 65°C for 5 minutes before placing directly on ice for 5 minutes. The RNA was then reverse transcribed in a final volume of 20 μl containing 50 mM Tris-HCl [pH 7.5], 75 mM KC1, 3 mM MgCl₂ , 10 mM dithiothreitol, 500 mM each deoxynucleoside triphosphate, and 40 U of Moloney murine leukemia virus reverse transcriptase. The reaction was allowed to proceed for 90 rnins at 37°C and the RNA:DNA hybrid products were then denatured at 90°C for 5mins and then chilled immediately on ice. A second strand synthesis reaction was then set up by addition of 200 ng UP-dN6, 8 units Klenow DNA polymerase, 5 μl 10 mM each dNTP, in a final volume of 50 μl. This reaction was allowed to proceed for 30 mins at 37°C. This procedure generates many double stranded DNA molecules with the core sequence of the UP-dN6 primer at both ends.

The products of this reaction were then purified using spin column chromatography (S-400™ column from Pharmacia, used as recommended) in order to remove unincorporated UP-dN6 primer. 5 μl aliquots of the eluted products were added to PCR reactions. PCRs were performed in 50 μl reaction volume containing 50 mM KCl, 10 mM Tris pH8.8 (at 20°C), 2 mM MgCl₂, 200 μM dNTPs, 2.5 units of Taq™ polymerase (Roche) and 10 pmol of UP oligonucleotide primer (5'- GCCGGAGCTCTGCAGAATTC). PCR conditions were 40 cycles of 94°C, 1 minute; 55°C 1 minute; 72°C 3 mins with an initial denaturation at 94°C for 4 mins. PCR products were analysed by electrophoresis through a 2% agarose gel to reveal a smear extending from around 200 bp up to lkb or larger. These products were then purified and blunt-end cloned into pBluescript™ (Stratagene), previously digested with EcoRY using standard methods.

Plasmid DNA was extracted from 19 individual colonies and the inserts sequenced using an Applied Biosystems™ automated DNA sequencer. Of these sequences obtained, 9 were identified as fragments of vesicular stomatitis virus (VSV) by database searches using he BLAST algorithm at the GenBank website (GenBank Accession code J02428). Therefore the virus infecting the original culture was identified as VSV using the shotgun sequencing approach. VSV is commonly used in many laboratories for pseudotyping other viruses and so was probably present as a laboratory contaminant of the original U87 culture. Following he molecular identification of this virus as VSV, this result was further validated by electron microscopy which clearly showed rhabdovirus particles in the culture and also by neutralisation of the infectivity using an anti-VSV antibody. 8 out of the other 10 sequences found were ribosomal sequences which may be expected since such sequences could be present in the viral pellet and be protected from nuclease digestion due to a high degree of secondary structure.

Although based on a relatively high titre virus in a relatively clean system this Example demonstrates the utility of the method of the invention in virus identification.

Example 2: Identification of putative viral sequences in pooled human serum

Factor VIII blood products for treatment of haemophiliacs are prepared from the pooled serum of many donors (up to 10,000). Virus particles present in human serum are therefore likely to be present in such blood products as was demonstrated by the inadvertent infection of many haemophiliacs with human immunodeficiency virus (HIV) in the early 1980s. Although blood products are now treated to eliminate any potential viruses being transferred to recipients, stocks of Factor VIII produced before this screening began still represent a potentially rich source of viral sequences. In addition, pooled serum from new donors without treatment to remove viruses also represents a.source of novel viruses. In this Example we used pooled factor VIII from paid donors (Zhang et al 1991 AIDS 5(6) 675-681).

To prepare a purified virus pellet, 500 μl of pooled factor VIII was first centrifuged at low speed (4000& 10 mins, 4°C) to remove cell debris. The supernatant from this spin was then re-centrifuged at high speed to pellet viral particles (100,000g, 15 minutes, 4°C). The viral pellet was then treated with micrococcal nuclease S7™ as described in Example 1 and DNA and RNA extracted. Fragments of DNA were amplified by sequence independent PCR essentially as described above except that, since this Example describes the identification of sequences of potential DNA viruses rather than RNA viruses, the sample was not treated with reverse transcriptase but instead had 2 cycles of extension with Klenow polymerase. The PCR products were cloned directly into pGEM-T-Easy™ (Promega) as recommended and transformed into XL 10 Gold™ E coli cells (Stratagene) as recommended. This bacterial strain is supplied as highly competent for transformation and its use therefore increases the size of the library produced.

Plasmid DNA from 77 clones containing inserts were selected from the library and sequenced using a Beckman-Coulter CEQ2000™ DNA sequencer. The sequences obtained were compared with database sequences. Any sequence which was not present in the human genome database nor in any other database of genomic sequences potentially represented a novel viral (or other pathogen) sequence. To date, two sequences have been obtained in this way which do not match the human genome database and whose absence from the human genome has been demonstrated empirically by the failure of specific PCR primers to amplify them from normal human genomic DNA.

Claims

1. A method for screening a biological sample from a human for the presence of a virus, comprising subjecting the sample to conditions which purify viral nucleic acid, shotgun cloning the purified viral nucleic acid, sequencing the cloned nucleic acid, and determining whether the sequence of the cloned nucleic acid is present in, or absent from, at least one human genomic database; wherein

- absence of the sequence from the database indicates that the cloned nucleic acid may be a viral nucleic acid.

2. A method according to claim 1 wherein the cloning includes a step of amplifying the nucleic acid by a sequence independent technique.

3. A method according to claim 2 wherein the nucleic acid is DNA and the .cloning includes amplifying the DNA by sequence independent PCR.

4. A method according to claim 2 wherein the nucleic acid is RNA and the cloning includes synthesising a complementary DNA strand by carrying out sequence independent reverse transcription and amplifying the DNA by PCR.

5. A method according to claim 3 or 4 wherein the sequence independent PCR or the sequence independent reverse transcription is carried out with a primer comprising a sequence NNNNNN wherein each N is A, T, G, or C.

6. A method according to any one of the preceding claims wherein the nucleic acid is cloned into a bacteriophage vector.

7. A method according to any one of claims 1 to 5 wherein the nucleic acid is cloned into a plasmid vector.

8. A method according to any one of the preceding claims wherein the step of subjecting the sample to conditions which purify viral nucleic acid includes treatment with a nuclease so as to digest extraneous nucleic acid, wherein the viral nucleic acid is protected from digestion by viral coat or core protein.

9. A method according to claim 8 wherein, prior to treatment with the nuclease, the viral nucleic acid is concentrated by centrifuging the biological sample under conditions such that cell debris is pelleted and virus particles remain in the supernatant; collecting the supernatant; and centrifuging the supernatant under conditions such that virus particles are pelleted.

10. A method according to any one of the preceding claims which further comprises determining whether the sequence of the cloned nucleic acid is present in, or absent from, at least one database comprising sequences of viruses.

11. A method according to claim 10 wherein the sequence of the cloned nucleic acid is absent from both the human genomic database and the database comprising sequences of viruses, thereby suggesting that the nucleic acid may be from an unknown virus..

12. A method according to claim 11 , which further comprises experimentally deteπriining whether the sequence of the cloned nucleic acid is present in human DNA.

13. A method according to claim 12 which comprises making a labelled probe comprising at least part of the sequence of the cloned nucleic acid and determining whether the probe hybridizes to human DNA.

14. A method according to claim 12 which comprises making a set of primers comprising sequence from the cloned nucleic acid and determining whether the primers are able to amplify human DNA.

15. A method according to any one of claims 11 to 14 which further comprises carrying out DNA walking to determine any sequence which flanks the sequence of the cloned nucleic acid.

16. A method according to any one of claims 11 to 15 which comprises isolating the unknown virus.

17. A method according to any one of claims 11 to 16 which further comprises determining whether the cloned nucleic acid or other nucleic acid of the same virus is present in a specimen from a patient who has a disease, wherein any presence of the nucleic acid in the specimen suggests that the nucleic acid may be from a virus which is causing or contributing to the disease.

18. A method according to any one of claims 11 to 16 which further comprises obtaining a specimen from each member of a group of subjects with a disease; determining whether the cloned nucleic acid or other nucleic acid of the same virus is present in each specimen; and determining whether the proportion of subjects in whom the nucleic acid is present is greater in the group of subjects who have the disease than in a control group of subjects who do not have the disease, wherein a said greater proportion suggests that the virus may cause or contribute to the disease.