WO2010016087A2 - Mutated sequences of hepatitis virus b related to drug resistance, method for their evaluation and use thereof in the medical field - Google Patents

Abstract

The invention relates to the identification of new mutations or combinations thereof (mutation profiles) in the inverse transcriptase gene and/or in the surface antigen of hepatitis B virus (HBV) potentially related with the genotypic resistance in vitro and in vivo to the therapy with antiviral drugs. In particular, the invention relates to methods for the genotypization of a viral population and for the subsequent correlation of such a genotypization to the fenotype features by an automated method, with the aim to bring to evidence the relationship between said mutations and the resistance to one or more antiviral agents. The invention also relates to methods for the use of the mutation profiles for diagnostic purposes and for the development of new drugs (drug design and modification of existing drugs), for the clinical treatment and for the definition of therapies to be administered.

Description

Mutated sequences of hepatitis virus B related to drug resistance, method for their evaluation and use thereof in the medical field.

Field of the invention

Object of the present invention are new mutated sequences of the hepatitis virus B related to drug resistance, a series of genomic analysis methods, statistics and artificial intelligence for the evaluation thereof and for use in the therapeutic field and a software implementation of said mathematic methods, in particular, the invention comprises the setting of an automated method for the evaluation of said mutations and for their correlation with antiviral drug resistance.

Definitions

Primary mutation = a mutation which is able to cause, alone, a reduced sensibility to a certain antiviral drug.

Compensatory mutation = a mutation which, even though it is not responsible on its own for the appearance of a reduced drug sensitivity, brings about a replicative advantage for a virus carrying a primary mutation, allowing in this way to increase the resistance level to the same drug.

Primer = a nucleic acid filament which acts like a starter point for DNA replication. Primers are necessary since many DNA-polymerases (enzymes which catalyze DNA replication) cannot start the synthesis of a new filament ex novo, but they can only add nucleotides to a preexisting filament.

Mutation profile = a combination of two or more mutations present at the same time in a sequence, also not adjacent to each other. Immune-escape mutants = viruses with mutations in the surface antigen which thus allow to escape the host's immune response and are responsible for the infection in subjects vaccinated against HBV.

Pyrosequencing (sequencing by synthesis) = an advanced genome sequencing technique based on the emission of a detectable signal as a consequence of the incorporation of a nucleotide in the nascent DNA chain and of the release of a pyrophosphate residue (Ronaghi et al. (1988) Science 281: 363.; Ronaghi et al. (1996). Analytical Biochemistry 242: 84.). The pyrosequencing device GS FLX manages automatically and in parallel a very large number of pyrosequencing reaction (massively parallel pyrosequencing).

Ultra-deep pyrosequencing = a sequencing carried out by a GS FLX pyrosequencing device on a large number of DNA fragments obtained by PCR sequencing of a small target region, which allows the detection of mutations present with a low frequency in the viral population analyzed.

Parsing or syntactical analysis = a process which enables the analysis of a continuous input flow (e.g. read from a file or a board) so that its grammatical structure can be determined thanks to a given formal grammar. A parser is a program which carries out this task. Usually parsers are not hand-written but are generated by parser generators. Tipically, the Italian term is used to refer to the acknowledgment of a grammar and to the subsequent building of a syntactic tree, which shows the rules used during the input acknowledgment; the syntactic tree is then visited (also more than once) during the work by an interpreter or by a compiler. In most languages, though, the syntactical analysis operates on a token sequence where the lexical analyzer breaks up the input. Thus, the English term is often used to indicate the whole of the lexical analysis and of the syntactical analysis itself.

Clustering, or cluster analysis, or grouping analysis = with these terms it is meant a group of multivariate analysis techniques of data able to select and group together the homogeneous elements in a set of data. All the clustering techniques are based on the concept of distance between two elements. Actually the robustness of the analyses obtained from the clustering algorithms essentially depends on the meaning of the metrics and, thus, on how the distance was defined. Distance is a fundamental concept, since the clustering algorithms group together the elements on the basis of their distance and for this reason belonging or not to a cluster depends on how distant a given element is from a cluster. The clustering techniques can possibly be based mainly on two approaches: a) Bottom-up. This approach envisages that all the elements are initially considered as clusters themselves and subsequently the algorithm pools the clusters that are nearest to each other. The algorithm keeps on pooling elements to a cluster until a set number of clusters is obtained or either until the minimum distance between clusters goes over a given value; b) Downwards. Initially all the elements are in one cluster only and subsequently the algorithm starts to divide the cluster into many smaller clusters. The principle which rules the division is always trying to obtain homogeneous elements. The algorithm proceeds until a set number of clusters is reached. This approach is also known as "hierarchic approach". Clustering techniques are usually used when a great number of heterogeneous data is to be searched for anomalous elements.

Machine learning = is one of the basic areas of artificial intelligence and is related to implementing systems based on observations or examples such as data for the synthesis of new knowledge, e.g., classifications, generalizations, re-formulations (Langley, P. (1996) "Elements of Machine Learning", Morgan Kaufmann editore; Mitchell, T. (1997). "Machine Learning", McGraw Hill editore; Witten, I. & Frank, E. (2005) "Data Mining: Practical Machine Learning Tools and Techniques", Morgan Kaufmann editor).

There exists a number of situations where traditional algorithms can not give easy answers. These typically are due to the presence of one or more factors, as follows:

- Difficulties in formalization,

- Great number of variable at play,

- Lack of theory,

- Personalization needs.

Automatic learning algorithms traditionally are divided into two main categories:

- Supervised learning: an instructor gives examples (and counterexamples) of what is to be learned;

Non-supervised learning: it starts from non-preclassified observations.

Prior Art

Despite the prevention done by vaccination, hepatitis B still is an important cause of mortality and morbidity. Currently, about 350 million people worldwide are chronically infected and, approximately up to a quarter of them develops a progressive disease ending up with cirrhosis and hepatocarcinoma (Fattovich G. Natural history of hepatitis (2003) B.J Hepatol; 39 (Suppl l):S50-8).

Hepatitis B virus (HBV) is a DNA virus which belongs to the hepadnaviridae family, with a circular incomplete double strand DNA genome of 3.2 Kb.

This virus developed a complex replication mechanism which involves formation of an intermediate called "covalently closed circular DNA" (cccDNA), which is transcribed by a DNA-dependent RNA-polymerase of the host in different RNA molecules. Among them the pre-genomic RNA is selectively packed within the capside and is subsequently retro- transcribed to DNA by the viral DNA polymerase (contained in the the capside itself) which is endowed with reverse transcriptase activity (rt). The infection by HBV is characterized by high replication levels (about 10¹¹ virions per day) e by a short half-life of the viral particles (1-3 days).

The enzyme responsible for the viral replication, the reverse transciptase, is not endowed with 3'-5' exonuclease activity and is thus not able to operate a correction of the translation errors made during its normal enzymatic activity. As a consequence, during every replication event, spontaneous mutations are introduced (up to 10¹⁰ per day). Consequently, HBV, like HIV and HCV, is present in an organism as a "quasispecies", i.e., a distribution of mutated, and sometimes recombinant, genomes which creates a population structure that is complex and dynamic. With the term "quasispecies" it is, therefore, meant a population of genetically distinct but very similar variants, generated by the casual accumulation of mutations in the replicating genomes, by a replication enzyme that does not work correctly, on which selection acts as an evolutionary mechanism of the quasispecies itself.

The virus has a 4 open reading frames genome which are partially overlapping; this genetic organization allows a mutation in a gene to often show up as a mutation in the overlapping gene, and this overlapping somehow decreases the setting of casual mutations in the viral quasispecies. The selective endogenous pressure (the immune system activity) and the exogenous one (the pharmacologic treatment) strongly influence the composition of the viral quasispecies, endowing the viral genome with a plasticity which is able to escape from the immune system action, and, possibly, from the antiviral drug action. The objective of the treatment of the chronic infection by HBV is the prevention of cirrhosis and hepatocarcinoma development, and also the prevention of hepatic transplantation failure.

Interferon (IFN)-α, a cytokine with immunomodulatory, antiviral and antiproliferative activity, was the first approved therapeutic agent, in 1990. During the last years the therapeutic approach has changed. Currently, in addition to interferons, many other drugs are available which have a direct antiviral activity, inhibiting HBV replicative activity (e.g., lamivudine, adefovir, entecavir and tenofovir).

Antiviral drugs currently available are all inverse transcriptase inhibitors (RTI), and are divided into nucleoside analogs (lamivudine, entecavir, telbivudine, emtricitabine, etc.) and nucleotide analogs (adefovir and tenofovir). Among these, lamivudine, adefovir, entecavir and telbivudine have already entered the current practice, while tenofovir, already used in patients co-infected by HIV and HBV as part of the anti-retroviral therapy, will soon be introduced also in the therapy of subjects infected by HBV only.

Nucleoside and/or nucleotide analogs are well tolerated in general but they call for very long treatment times, virtually for the duration of life. Their choice is a compulsory one in those patients which are intolerant to the IFN therapy, in non-responders and in those cases where the clinical condition is very advanced (patients with advanced cirrhosis), or with complications (patients subjected to chemiotherapy and immunocompromised patients) for whom it is necessary to proceed with a rapid dampening of the viral replicative activity. In these patients the treatment of choice is the one with nucleoside/nucleotide analogs.

These drugs block the HBV inverse transcriptase activity and thus by inhibiting the viral replication within the infected cells. Unfortunately, one of the most relevant issues of the treatment with these drugs and, more generally, of the nucleoside and/or nucleotide analogs, is related to the appearance of resistant mutants, a phenomenon which is made easier by the sloppiness of the viral replicative enzyme.

Resistance can be evaluated in terms of genotypic resistance (appearance of mutations associated with drug resistance), virologically (increase in the viral DNA levels in treated patients sera) and biochemically (increase of transaminases). Tipically, the increase in viral load of at least one order of magnitude is one of the events used to define the treatment failure in virological terms.

The therapeutic experience with antiviral drugs in case of HBV infection shows that the resistance to an antiviral drug is gradually acquired through the selection of pre-existent variants in the viral quasispecies, which show mutations in certain positions of the inverse transcriptase gene. The progressive increase of resistance to a certain drug is favored by the gradual accumulation of mutations which endow the virus with a survival advantage, and, therefore get fixed in the viral quasispecies.

Assays for the detection of the HBV genomic mutations are based on the direct sequencing of the reverse transcriptase (rt) gene, after specific viral sequences amplification by the polymerase chain reaction (PCR).

The latter approach, based on methods developed in home by each laboratory or on commercially available methods, such as for example the Trugene HBV Genotyping kit system (Siemens Medical Solutions Diagnostics), allows the identification of all the possible mutations (either primary or compensatory) already known, or newly identified, present in the amplified region, but is limited by not being able to detect some mutations in a minority of the circulating viral quasispecies, since it generally detects the sequence(s) present with a frequency of at least 20% of the total virus population (Lok, Zoulim, et al (2007) Hepatology).

In addition to the direct sequencing, there exist other techniques able to yield a more sensitive response. Among these, a very used technique to carry out a genotypic resistance test for HBV is reverse hybridization (reverse hybridization line-probe assay, LiPA) (Stuyver, Van Geyt et al. (2000) J Clin Microbiol; Kim, Han et al. (2005) Antiυir Ther; Hussain et al (2006); Osiowy (2006); Libbrecht et al (2007)) which is used in the commercial kit INNO-LiPA HBV DR v2 (Innogenetics NV, Ghent, Belgium). After amplification is carried out, the DNA is hybridized with specific oligonucleotide probes immobilized onto the membrane of the strip supplied with the kit.

This method allows the detection of drug-resistant viral variants in the total quasispecies with a low frequency, also in the order of 5%. In general, this technique is more simple, quicker and more sensitive with respect to the direct sequence, but is severely limited as it is able to detect only mutations which are already known for which specific probes have already been created. In particular, the commercial test DRv2 can detect simultaneously the wild type (wt) variant and the variants which carry the mutations or the polymorphisms of codons 80, 173, 180, 181, 204, and 236 only of the HBV rt gene.

The HBV reverse transcriptase, like HIV and other viruses, contains seven well conserved regions (A-E). A, C and D regions constitute the linking domain between the catalytic core and the nucleotide precursors (dNTP), domain B is involved in the bond of RNA or DNA molecule to be copied (template) and domain E is linked to the nucleotide chain portion which act as a template for the replication (primer strand, Bartholomeusz A. (2004), das K 2001). The mutations most frequently related to the lamivudine resistance concern the YMDD motif (tyrosine, methionine, aspartic acid, aspartic acid) situated in the C domain of reverse transcriptase, with a substitution of valine or isoleucine for methionine (M204V/I) (Allen MI, Deslauriers M. et al. Lamivudine Clinical Investigation Group. Hepatology (1998);27:1670-1677.). More recently, in this same position a more rare mutation, from methionine to serine (M204S) (Bozdayi AM, et al. (2003) J Viral Hepα£.;10:256-65) was identified.

The lamivudine resistant mutants with the mutated M204 position replicate less efficiently than the wild type virus in vitro, especially when the mutation is from M to V (20). Therefore, in lamivudine resistant patients, the M204V mutation (and, less importantly, the M204I one) is accompanied most often by compensatory mutations, which may effectively restore the replication ability of the virus. The L108M (18, 22) mutation and the V173L mutation in the B domain (18, 21, 22), are among the most frequent compensatory mutations (9-22% of the cases). Other compensatory mutations, which progressively accumulate during the course of the lamivudine treatment are: L80V/I, L82M, F166L, A200V and V207I (23-26).

Given the similarity of the therapeutic target for the the nucleoside and nucleotide antiviral drugs (reverse trnscriptase), the mutations able to confer resistance to one drug often confer resistance to another drug, especially when they belong to the same class. In view of the continuous evolution of the therapeutic antiviral paraphernalia against HBV, the literature gets continuously enriched by new information about this phenomenon, which is called cross-resistance.

The identification of mutations linked to the lamivudine resistance is very important, also with regard to the use of alternative drugs, above all with reference to the availability of new drugs against lamivudine- resistant strain.

Adefovir is a nucleotide analog approved for the hepatitis B treatment. This drug is active against lamivudine-resistant strains. Resistance against adefovir develops much more slowly than resistance against lamivudine. In any case, at the end of a five-year treatment the percentage of patients with mutations gets to 28%. The main mutation associated with adefovir resistance is N236T in the D domain of polymerase; more recently other mutations heve been associated with adefovir resistance, such as the A181T/V in the B domain, the V84M and the S85A in domain A and the L217R and the I233V in the D domain. All these changes have been associated with a reduced sensitivity to the drug in vitro.

Entecavir is a potent drug which has lately entered the clinical practice for the chronic hepatitis treatment. In a trial carried out in 181 patients treated with entecavir for 90 weeks, of which 87% was lamivudin resistant, the following combinations of mutations were observed: I169T+L180M+M204V+M250V and

L180M+M204V+T184G+S202I. These results suggest that the entecavir resistance needs the build-up of four mutations to show up, comprising two involved in lamivudine resistance (Tenney DJ et al. (2004) 48:3498-3507).

Experience made in the last years taught that nutations developed in response to a drug can influence the therapeutic response to a second drug, preventing its full effectiveness. More recent experiences about therapeutic failures also led to a change in the approache to rescue therapies. As a matter of fact, in case of patients with a previous therapeutic failure it was observed that it is preferable to add a second drug instead of substituting the antiviral drug previously used. Precociously spotting mutations associated with a resistance and defining the pattern of the mutations, which sometimes can be very complex, is important for the choice of the most suitable therapy in each single case. In case of patients never treated before there may, in addition, be present minor pre-therapy variants in the viral population which could influence the therapeutic response.

In the attempt to obviate these problems, a genotypic analysis is usually carried out on the HBV populations found in a patient, using the known techniques described above, trying to correlate the mutations possibly found with the resistance data present in the literature about this subject. Anyway, it is clear to the expert in the art that the latter task is practically impossible to be carried out using the tools currently found in the state of the art, since the information is very unorganized and fragmentary. Indeed, there is not such a database, i.e., that gathers information about all the possible mutations and about their significance with respect to a potential therapeutic approach. Therefore, the current state of the art does not allow the specialist physician to select an antiviral drug or a combination of such drugs with a good probability that the therapy against HBV would be effective.

In addition, there does not exist an automated method which, supported by a genotype profiling, allows the subsequent correlation of such a profile to fenotype features with the aim to emphasize the relationship between said mutations and the resistance to one or more antiviral agents. Such an automated method should respond to certain criteria of input rates and processing thereof which the tools used in the field of diagnosis and therapy of the HBV infection are not currently able to provide.

There also are not reliable, rapid and easy methods to be implemented which allow the identification of mutations and/or mutation profiles occurring at <20%.

Also, in the prior art there are not rapid method for the use of mutation profiles of the HBV either for drug design purposes and modification of existing drugs, and for diagnostic and clinical purposes.

Finally, there does not exist a rapid method which, based on sequence data of rt gene of HBV, gives data referring to the sequence of the overlapping gene which codifies for the surface glycoprotein. The mutations of this antigen are important since they could alter the antigen recognition by the immune system, allowing the escape from the defensive immune mechanisms and a poor detectability by the serologic methods for HBsAg determination, which is a viral infection and replication marker.

The possibility to decide a priori which drug can be administered to a patient with a good chance that the therapy is effective, is crucial, since it would help avoiding exposure of patients to unnecessary side effects and to useless expenses. As already described previously, mutations or combinations thereof (mutation profiles) related or which can be related to antiviral drugs are already known. But many others are continuosly discovered whose role is still to be conclusively confirmed, and it is likely that new mutations will be identified in the future, whose role, as well, is to be established. It is therefore considered useful and necessary to set up a method by which it would be possible to store all the data about the clinical significance of this mutations and to make them correlate with ease and speed to the therapy to be administered to a certain patient.

Summary of invention

It has been now found and is the object of the present invention an automated method which allows the manipulation, i.e., the evaluation, the analysis and management of a large number (in the order of tens of thousand) of sequences coming from traditional sequencing and from ultra-deep sequencing of HBV viral genome. The manipulation required comprises processing, parsing, alignment, error correction, clustering, phylogenetic analysis and data variation, extraction of the reverse transcriptase and/or of the surface antigen of HBV virus gene mutations from such sequences. The method additionally allows evaluating of the nucleotide and aminoacid variability and the consequent extraction of significant mutations and/or mutation profiles. The method also advantageously comprises a further stage of correlating the mutations found to the drug treatment resistance, in particular antiviral drugs, of at least one strain from the mutant viral isolate.

A further object of the invention is the device for the implementation of the different stages of the automated method to obtain the mutations and/or mutation profiles of HBV.

Another object of the invention are the mutations of the HBV genome detected by the automated method of the invention and shown in Table 1 and their mtation profiles.

Still another object is the use of the pyrosequencing technique for the genotyping of the viral population present in a patient infected with HBV.

A further object is the use of said mutations and/or mutation profiles to build a database of the genomic mutations of HBV to be used in the medical field, in particular for the diagnosis and treatment of drug resistant strains of HBV. Such a database can be implemented with the results of sensitivity tests to drugs carried out in vitro (phenotypical tests) and in the clinical setting (in, vivo) for use in combination with the automated method of the invention.

Still a further object is a method for the evaluation of antiviral therapy efficacy in a patient infected by HBV, being said method carried out in a biological sample, preferably blood, from the patient, and comprising the stages of: determining if the sample comprises at least a nucleic acid which codifies for the reverse transcriptase and/or the surface antigen of HBV containing at least one mutation linked to resistance; and using the presence of said mutated nucleic acid to evaluate the efficacy of the antiviral therapy.

Still a further object of the invention is a method for determining the sensitivity of the viral population in a patient infected with HBV to one or more drugs, being said method carried out on a biological sample, preferably blood, of the patient, and comprising the stages of: determining if the sample comprises at least one nucleic acid that codifies for the reverse transcriptase and/or the surface antigen if HBV containing at least one mutation linked to resistance; and use the presence of said mutated nucleic acid to choose an antiviral therapy.

It is still an object of the invention a method to identify drugs effective on HBV strains resistant to therapy with nucleotide and/or nucleoside antiviral drugs, said method comprising the stages of: obtaining at least one HBV wild type or engineered strain comprising in its genome at least one of the mutations identified with the method of the invention; detecting the phenotypic response to the assay drug; and using the phenotypic response thus obtained to determine the efficacy of the tested drug.

Still further objects of the invention are: oligo- or polynucleotide sequences containing the mutations and/or the mutation profiles generated according to the method of the invention to the method of the invention to be used in the medical field, in particular for diagnosis and therapy purposes; plasmids comprising said oligo- and polynucleotides to be used as cloning and expression vectors; engineered HBV viruses containing one or more mutations resistant to antiviral drugs.

Further objects of the invention will be evident from what described herein below.

Brief Description of the Figures

Figure 1: block diagram of the operating scheme of the automated method of the invention.

Figure 2: frequency of the mutated positions in the rt of HBV genotype A or D, obtained by applying the ultradeep pyrosequencing by GS FLX to biological samples from patients.

Detailed description of the invention

The present invention relates to the field of nucleic acid-based diagnosis and to the detection of mutations in certain nucleic acid sequences. In particular, the invention relates to a method for the genotypization of the viral population in an HBV infected patient carried out by a system able to analyze the complexity of the viral quasispecies, also allowing detection of minor viral populations, tipically <20%, preferably <15%, more preferably <10%, most preferably <5%, and to correlate them to one or more antiviral agents.

The invention relates in particular to an automated method which allows analysis of the results from massive sequencing, in addition to correlation of the mutations thus detected (possibly present in minor viral populations, thus not easily detected with the traditional direct sequencing methods) to their phenotype characteristics with the aim to make evident the relationship between said mutations and/or mutation profiles and resistance, with no need for an expensive and difficult to apply in loco phenotypization of the mutants observed.

The mutations and/or mutation profile thus obtained constitute a database thet can be used for diagnosis, for the clinical treatment and in the setting of therapies.

The method of the invention uses an approach which combines checking for the presence of a certain mutation and/or combinations of mutations in the genome of an HBV population in a patient, detected by any sequencing method, but preferably with the so called ultradeep pyrosequencing, and an in vitro check, by a phenotypic test, of the actual effect of the detected mutation, on the resistance to one or more antiviral drugs, or the in vivo assay with reference to clinical paramenters, such as, for example, the viral load.

As already mentioned above, the genotypic test of the HBV resistances may comprise a sequencing stage of the fragments of the viral amplified genome, which is tipically carried out according to the Sanger method (Sanger F, Coulson AR. (1975) JMoI Biol. 94:441-448.) Preferably, the method of the invention takes advantage of the depth of analysis of a new DNA sequencing system called "ultra-deep pyrosequencing", which is carried out by means of the latest generation sequencing system called GS FLX. Such a sequencing device has been applied to the sequencing of whole eucariotic genomes (Margulies et al. (2005) Nature 437:376-380) and recently it is finding new applications in microbiology and virology.

Pyrosequencing is a relatively recent DNA sequencing technique based on synthesis instead of chain interruption according to the classic method of Sanger. It was originally developed at the Royal Institute of Technology in Stockholm in 1990 (Nyren, P. (2007). Methods MoI Biology 373: 1-14).

Pyrosequencing is based on 5 steps:

1) the sequence to be analyzed, after PCR amplification, is denatured and incubated as a single helix with the DNA polymerase enzymes, ATP solforilase, luciferase and apyrase and with the substrates adenosinesulfophosphate (ASP) and luciferin;

2) one of the four dNTPs is added to the reaction. The DNA polymerase catalyzes the addition of such a base only if it is complementary to the residue on the template. In this case, inorganic pyrophosphate (PPi) is developed concomitantly;

3) the PPi thus produced is transformed into ATP by the sulforylase using ASP a a substrate. The ATP thus obtained allows the conversion of luciferin into oxyluciferin to happen by luciferase with the appearance of a light signal which is detected by a fluorimeter;

4) the enzyme apyrase degrades the dNTP that was not incorporated and the ATP produced by the sulforylase. Only when the degradation is finished a second dNTP is added to make the polymerization reaction go further (going back to step 1);

5) all the four dNTPs are added cyclically until the complete sequence is deduced, the light signal produced each time by luciferin is registered in a "pyrogram". The signal will be proportional to the ATP produced and, therefore, to the number of nucleotides of the same type which have been added to the nascent DNA chain.

The GS FLX system, in particular, takes advantage of the pyrosequencing reaction described above and of a miniature detecting system. Thousands DNA fragments are adsorbed to a solid support constituted by microbeads (one fragment for each bead). The microbeads are suspended in an emulsion so that each is contained in one single emulsion micelle, where a PCR reaction is carried out which amplifies the single adsorbed fragment (emulsion PCR). The beads, charged with the amplified fragments, are then distributed in a chamber which contains thousands of cells, in which separate pyrosequencing reactions takes place (one for each bead). The signal production in each single cell is registered by a very high resolution CCD camera system, which produces a measurement for each single cell. The sequence report comprises thousands of sequences, one for each single initial fragment (ultra-deep pyrosequencing) of viral DNA, thus allowing the detection of rare sequences (below 20%) within the viral quasispecies.

The fenoypic assay of the resistances is carried out to determine the ratio between the drug dose that inhibits in vitro by 50% the mutated virus replication (IC50) and the IC50 of the wt reference virus in systems based on cell cultures. The test is based on the introduction of the genome fragment coming from the patient and containing the mutation(s) in an HBV genomic background able to replicate in vitro. The recombinant virus thus constructed is introduced into a cell system able to replicate it (tipically HepG2 cells) and the cell culture is carried on for a few days in the presence or absence of scalar doses of the drug to which the sensitivity is to be tested. The viral replication is measured by dot blot or southern blot using radioactively labelled probes, by means of which the number of the produced genomes is quantified. Based on the comparison between the cell culture without the drug and the one in the presence of the drug at the different concentrations an inhibition curve is inferred on which the concentration value of the drug able to give an inhibition of 50% (IC50) of the drug production is interpolated. A viral strain is defined as resistant to a certain drug when an increase of at least 10 times of the IC50 is observed. In principle, it would not be necessary to know a priori the mutation present in the virus for the fenotypic resistance test to be performed.

The basic methodology to obtain HB viruses artificially mutated is similar to the one also applied to other types of virus and has already been described (see Zoulim F. (2006) Semin Liver Pis. 26:171-180) even though tests for HBV are not commercially available yet.

A very interesting aspect of the development of resistant mutants is linked to the fact that the polymerase gene overlaps that of the surface antigen. It is possible that multiple mutations in the polymerase accumulate in patients undergoing treatments with nucleotide/nucleoside analogs, which may cause simultaneous changes on the surface antigen (HBsAg), causing the selection of "immune escape" variants that escape from the immune system control of the host. These mutations, which are difficult to detect with the traditional methods known in the art, can be detected with the method of the invention as well.

In one of its preferred embodiments, the present invention relates to the new mutations of Table 1, as compared to the HBV wild type genotypes D and A. Mutations are indicated with the code x_n_y and selected from:

Table 1

s:E_2_A; s:E_2_D; rt:D_2_G; rt:W_3_R; s:N_3_S; rt:G_4_E; rt:G_4_R; s:I_4_T; s:T_5_A; s:T_5_I; s:T_5_S; rt:A_7_D; s:G_7_R; rt:A_7_T; rt:A_7_V; rt:A_7_Y; rt:E_8_D; s:F_8_H; rt:E_8_K; s:F_8_P; s:F_8_R; s:L_9_R; rt:H_9_Y; s:G_10_A; s:G_10_K; rt:G_10_R; s:G_10_R; rt:E_ll_D; rt:E_ll_K; s:P_ll_L; rt:E_ll_Q; s:L_12_V; rt:H_13_L; rt:H_13_N; s:L_13_P; rt:H_13_R; rt:H_13_Y; s:V_14_A; s:V_14_G; rt:R_15_K; s:G_18_del; rt:R_18_K; rt:R_18_S; s:F_19_C; s:F_20_L; rt:P_20_S; s:F_20_S; s:L_21_S; rt:R_22_C; rt:R_22_G; rt:V_23_A; rt:T_24_A; s:R_24_K; s:R_24_L; s:R_24_S; s:I_25_T; rt:G_26_R; s:T_27_del; rt:V_27_F; s:T_27_I; s:I_28_del; s:I_28_L; s:P_29_del; s:P_29_L; rt:V_30_G; s:Q_30_H; s:Q_30_K; s:Q_30_W; rt:D_31_N; s:S_31_N; s:S_31_R; s:L_32_Q; s:D_33_A; s:D_33_L; s:S_34_L; s:W_35_F; s:W_35_G; rt:H_35_Y; rt:N_36_H; s:W_36_L; s:W_36_U; s:T_37_P; rt:T_37_S; s:T_37_S; rt:A_38_E; rt:A_38_P; rt:A_38_S; rt:A_38_T; s:L_39_D; s:L_39_del; s:L_39_P; s:N_40_del; s:N_40_F; s:N_40_H; s:N_40_S; s:N_40_T; s:F_41_L; s:L_42_G; s:L_42_R; s:G_43_del; s:G_43_M; s:G_43_R; s:G_44_E; s:E_44_R; s:S_45_A; s:T_45_A; s:S_45_ins; s:T_45_N; s:T_45_S; s:S_45_V; s:P_46_G; s:T_46_P; s:P_46_V; s:V_47_A; s:V_47_del; s:V_47_E; s:V_47_G; s:V_47_I; s:V_47_ins; s:V_47_K; s:V_47_N; s:V_47_T; s:C_48_F; s:C_48_G; rt:Q_48_H; s:L_49_P; s:L_49_R; s:G_50_A; s:G_50_S; rt:S_50_W; rt:R_51_K; s:Q_51_L; s:Q_51_R; s:Q_51_V; s:N_52_D; rt:G_52_E; rt:G_52_R; rt:N_53_D; rt:N_53_I; rt:N_53_K; s:S_53_L; rt:H_53_S; rt:N_53_S; rt:N_53_V; rt:H_54_D; s:Q_54_H; s:Q_54_N; s:Q_54_R; rt:H_54_S; rt:H_54_T; rt:R_55_del; s:S_55_del; s:S_55_F; rt:R_55_H; rt:R_55_Q; rt:V_56_M; s:P_56_Q; s:P_56_S; s:P_56_T; s:T_57_A; s:T_57_I; s:T_57_S; s:S_58_C; s:S_58_F; rt:W_58_U; s:N_59_S; s:N_59_Y; s:H_60_Q; s:H_60_R; s:S_61_L; s:S_61_P; s:P_62_L; s:T_63_I; s:T_63_N; s:T_63_S; s:S_64_C; s:S_64_P; s:C_65_Y; s:P_67_Q; s:T_68_I; s:P_70_A; s:P_70_L; s:W_74_U; s:M_75_L; s:C_76_F; s:C_76_S; s:C_76_Y; rt:S_78_C; s:R_78_Q; s:R_79_C; s:R_79_G; s:R_79_H; s:R_79_L; s:R_79_S; s:F_80_C; rt:L_80_I; s:F_80_L; s:F_80_S; rt:L_80_V; s:I_81_H; s:I_82_L; rt:L_82_M; s:I_82_M; s:F_83_C; s:F_83_L; s:F_83_S; s:L_84_P; s:F_85_C; s:F_85_I; s:I_86_R; rt:A_87_E; rt:A_87_G; rt:A_87_T; rt:A_87_V; rt:F_88_S; rt:Y_89_S; rt:H_90_P; s:C_90_S; rt:I_91_F; rt:I_91_P; rt:I_91_V; s:I_92_L; s:I_92_M; rt:L_93_N; s:V_96_G; s:L_97_del; rt:A_97_S; rt:A_97_T; s:L_98_V; s:S_100_C; s:Y_100_C; s:Q_101_R; rt:V_103_I; s:M_103_I; s:M_103_L; rt:I_103_V; rt:G_104_S; s:P_105_A; s:P_105_S; s:A_105_V; rt:S_106_A; s:V_106_G; s:V_106_I; s:C_107_D; rt:G_107_R; s:C_107_R; s:C_107_Y; s:L_109_G; rt:S_109_P; rt:R_110_G; s:I_110_L; s:I_110_V; s:Pllll_L; s:P_lll_R; rt:V_112_A; s:G_112_A; s:G_112_E; rt:V_112_L; s:G_112_R; s:S_113_A; s:S_113_del; rt:A_113_G; s:S_113_T; rt:R_114_H; s:S_114_P; s:S_114_T; s:T_115_P; rt:L_115_V; s:T_116_S; s:S_117_R; s:S_117_T; s:T_118_A; rt:N_118_D; rt:N_118_S; rt:N_118_T; s:T_118_V; rt:N_118_Y; rt:R_120_K; rt:R_120_S; s:P_120_S; s:P_120_T; rt:I_121_N; rt:I_121_S; rt:I_121_V; rt:F_122_H; rt:F_122_I; s:R_122_K; rt:F_122_N; s:R_122_Q; rt:F_122_S; rt:F_122_V; rt:F_122_Y; rt:N_123_D; rt:N_123_T; s:T_123_V; rt:Q_125_H; s:T_125_M; s:T_126_I; s:T_126_S; rt:H_126_Y; s:P_127_L; rt:G_127_R; s:P_127_T; rt:T_128_A; s:A_128_E; rt:T_128_I; rt:T_128_N; s:A_128_T; s:A_128_V; s:Q_129_H; rt:M_129_L; s:Q_129_R; s:Q_129_U; s:G_130_del; s:G_130_E; s:G_130_R; s:N_131_A; s:T_131_A; rt:D_131_H; s:T_131_N; rt:D_131_S; s:K_133_I; s:M_133_I; s:K_133_L; s:M_133_R; s:K_133_T; s:M_133_T; s:F_134_C; s:Y_134_C; s:F_134_del; rt:D_134_E; s:Y_134_F; rt:D_134_G; rt:D_134_N; s:Y_134_N; rt:D_134_V; s:P_135_A; s:P_135_H; rt:H_135_T; s:S_136_L; rt:C_136_Y; s:C_137_G; rt:S_137_T; s:C_138_G; rt:R_138_K; rt:D_139_H; rt:D_139_K; rt:D_139_Q; rt:D_139_S; s:C_139_Y; s:T_140_del; s:I_140_S; s:T_140_S; s:K_141_del; rt:Y_141_F; rt:Y_141_Q; s:K_141_T; s:K_141_U; rt:V_142_E; rt:V_142_I; s:S_143_L; s:T_143_L; s:S_143_T; s:G_145_E; rt:L_145_M; s:G_145_R; s:N_146_H; s:N_146_T;s: C_147_Y; s:T_148_del; rt:Y_148_F; rt:K_149_H; s:C_149_R; s:C_149_Y; s:I_150_T; rt:F_151_L; s:P_151_L; rt:F_151_V; rt:F_151_Y; s:I_152_T; s:I_152_V; rt:R_153_Q; rt:W_153_R; rt:R_153_W; s:S_154_A; rt:K_154_E; rt:K_154_N; s:W_156_R; rt:L_157_P; s:F_158_L; s:G_159_A; s:G_159_E; rt:S_159_T; rt:H_160_R; s:K_160_R; s:Y_161_L; s:F_161_Y; s:L_162_I; s:W_163_C; s:W_163_U; rt:I_163_V; s:E_164_A; s:E_164_D; s:E_164_G; rt:L_164_M; s:E_164_V; s:L_165_U; s:A_166_V; s:V_168_D; rt:K_168_E; s:A_168_V; s:R_169_G; s:F_170_ins; s:F_170_S; s:F_170_T; s:W_172_C; rt:G_172_U; s:W_172_U; s:L_173_F; rt:V_173_L; s:L_173_P; s:S_174_N; s:S_174_T; s:L_175_F; s:L_175_S; s:L_176_R; s:V_177_A; s:V_177_E; s:V_177_M; s:P_178_Q; s:P_178_T; s:F_179_L; s:F_179_V; s:V_180_A; s:V_180_L; rt:L_180_M; s:Q_181_E; rt:A_181_S; rt:A_181_T; rt:A_181_V; s:W_182_U; s:F_183_C; s:F_183_L; s:V_184_A; rt:T_184_S; rt:A_186_T; s:S_187_del; rt:I_187_L; rt:I_187_T; rt:I_187_V; rt:C_188_R; s:T_189_I; s:V_190_A; s:V_190_G; s:W_191_U; s:L_192_F; s:L_192_I; rt:R_192_S; s:L_192_V; s:S_193_K; s:S_193_L; s:S_193_N; s:V_194_A; s:A_194_D; s:A_194_G; s:I_195_L; s:I_195_M; s:I_195_T; s:W_196_C; s:W_196_L; s:W_196_S; s:W_196_U; s:M_197_I; s:M_197_L; s:M_197_T; s:M_198_I; s:M_198_K; s:M_198_R; s:M_198_T; s:W_199_R; s:W_199_U; rt:L_199_V; s:Y_200_F; rt:A_200_G; s:Y_200_L; s:Y_200_N; s:Y_200_S; rt:A_200_T; rt:A_200_V; s:W_201_del; rt:F_201_U; s:W_201_U; s:G_202_A; rt:S_202_R; s:G_202_R; s:P_203_ins; s:P_203_L; s:P_203_Q; s:P_203_R; rt:Y_203_S; s:P_203_T; s:P_203_U; s:R_204_G; s:S_204_G; rt:M_204_I; s:R_204_K; rt:M_204_L; s:R_204_N; s:S_204_N; s:S_204_R; rt:M_204_V; s:L_205_G; rt:D_205_N; rt:D_205_Y; s:Y_206_C; rt:D_206_E; s:Y_206_F; s:Y_206_H; s:Y_206_P; s:Y_206_Q; s:Y_206_R; rt:D_206_U; s:R_207_C; rt:V_207_E; s:R_207_G; s:N_207_H; rt:V_207_L; rt:V_207_M; s:R_207_N; s:N_207_R; s:N_207_T; rt:V_208_A; rt:V_208_E; s:I_208_N; s:I_208_S; s:I_208_T; s:L_209_I; s:L_209_M; s:L_209_T; rt:L_209_V; s:L_209_V; rt:G_210_E; s:R_210_K; s:S_210_M; s:R_210_N; s:S_210_N; s:S_210_R; s:R_210_T; rt:A_211_D; rt:A_211_I; s:P_211_L; s:P_211_Q; s:F_212_del; s:F_212_L; rt:K_212_N; rt:K_212_R; s:I_213_F; s:L_213_I; s:I_213_M; s:L_213_M; rt:S_213_R; s:I_213_S; s:L_213_T; rt:V_214_A; s:P_214_H; s:P_214_L; rt:P_215_H; rt:P_215_T; s:L_216_C; s:L_216_F; rt:H_216_Q; s:L_216_U; rt:H_216_Y; s:L_217_E; rt:L_217_H; s:L_217_ins; rt:L_217_R; rt:E_218_D; rt:E_218_H; s:I_218_ins; s:I_218_S; s:I_218_T; s:F_219_I; s:F_219_S; s:F_220_C; s:F_220_del; rt:L_220_F; s:F_220_L; s:F_220_S; rt:L_220_T; s:C_221_G; s:C_221_R; rt:F_221_Y; s:C_221_Y; rt:T_222_A; s:L_222_del; s:L_222_P; rt:T_222_S; s:L_222_V; rt:A_223_P; s:W_223_R; rt:A_223_S; s:W_223_U; s:V_224_A; s:V_224_E; s:V_224_G; rt:V_224_I; rt:V_224_L; s:Y_225_C; s:Y_225_F; s:Y_225_H; rt:T_225_P; rt:T_225_S; s:Y_225_S; s:Y_225_U; rt:N_226_H; rt:F_227_H; rt:L_228_P; rt:L_229_M; rt:L_229_S; rt:L_229_V; rt:L_229_W; rt:S_230_C; rt:S_230_P; rt:S_230_Y; rt:L_231_H; rt:L_231_S; rt:G_232_V; rt:I_233_L; rt:I_233_M; rt:I_233_P; rt:I_233_T; rt:I_233_V; rt:H_234_L; rt:L_235_F; rt:L_235_U; rt:L_235_V; rt:N_236_K; rt:N_236_T; rt:N_236_V; rt:P_237_L; rt:P_237_Q; rt:P_237_T; rt:A_238_H; rt:A_238_K; rt:A_238_S; rt:K_241_R; rt:R_242_G; rt:G_244_R; rt:Y_245_H; rt:S_246_N; rt:S_246_P; rt:S_246_T; rt:N_248_H; rt:H_248_Y; rt:F_249_L; rt:M_250_I; rt:M_250_T; rt:Y_252_H; rt:V_253_I; rt:I_253_T; rt:G_255_R; rt:S_256_C; rt:C_256_G; rt:Y_257_G; rt:Y_257_H; rt:W_257_L; rt:Y_257_W; rt:S_259_L; rt:T_259_P; rt:T_259_S; rt:S_259_T; rt:Q_262_U; rt:H_264_Y; rt:I_266_T; rt:I_266_V; rt:Q_267_H; rt:Q_267_L; rt:Q_267_R; rt:Q_267_Y; rt:I_269_L; rt:I_269_T; rt:E_271_A; rt:E_271_D; rt:H_271_D; rt:H_271_E; rt:E_271_H; rt:E_271_L; rt:E_271_Q; rt:R_274_G; rt:R_274_Q; rt:R_274_S; rt:I_278_A; rt:V_278_I; rt:R_280_M; rt:D_283_E; rt:W_284_R; rt:K_285_Q; rt:V_286_L; rt:C_287_F; rt:Q_288_U; rt:R_289_S; rt:I_290_L; rt:I_290_N; rt:V_291_G; rt:V_291_M; rt:V_291_W; rt:G_292_A; rt:G_292_W; rt:L_293_V; rt:L_294_F; rt:L_294_U; rt:G_295_D; rt:F_296_L; rt:F_296_M; rt:F_296_S; rt:A_297_I; rt:A_297_T; rt:A_298_D; rt:A_298_V; rt:F_300_Y; rt:T_301_K; rt:Q_302_E; rt:C_303_F; rt:Y_305_A; rt:P_306_S; rt:P_306_T; rt:A_307_V; rt:L_308_E; rt:K_309_L; rt:P_310_L; rt:L_311_S; rt:A_313_T; rt:Q_316_H; rt:A_320_P; rt:F_321_T; rt:T_322_P; rt:T_322_S; rt:T_322_V; rt:F_323_L; rt:S_324_R; rt:P_325_H; rt:A_329_D; rt:A_329_T; rt:F_330_C; rt:L_331_V; rt:C_332_F; rt:C_332_N; rt:C_332_R; rt:C_332_S; rt:C_332_Y; rt:K_333_N; rt:K_333_T; rt:Y_335_F; rt:L_336_M; rt:N_337_D; rt:N_337_H; rt:N_337_T;

wherein: rt = reverse transcriptase gene (polymerase); s = gene codifying for the surface antigen; n = positive integer indicating the position of the aminoacid codon

(standard numbering); x = aminoacid in wild type (D or A genotype); y = aminoacid substituting the wild type one (mutation); in addition, when y = U the mutation introduces a stop codon.

The mutations listed in Table 1 can be present as single mutations or in association, and can confer to HBV resistance to treatment with at least one antiviral drug as far as the reverse transcriptase (rt) is concerned, and an immune escape phenotype as far as HBsAg is concerned.

The sequences and the mutations of the invention were obtained as follows: pyrosequencing from 13 blood samples taken from 8 patients treated with lamivudine/emtricitabine or with lamivudine followed by adefovir, and from 5 naϊve patients; downloading of the HBV database from GenBank; extracting mutations from wild type genotypes D or A from the above mentioned sequences, with optimized local alignment and error correction; evaluating the nucleotide and aminoacid variability (with calculation of the confidence interval on the mutation frequencies); extracting the mutations with frequency above 1%.

Further investigation was carried out to identify, by a statistical method, the mutations which were positively associated in vivo to the antiviral treatment. We collected HBV RT sequences from 336 treated patients. The treatment administered to these subjects were divided as follows: about half were receiving lamivudine as a monotherapy; the remaining patients, in about equal proportions, were receiving adefovir and lamivudine as a combination, tenofovir and lamivudine as a combination, entecavir as a monotherapy, and a a multiple alignment was performed, translating into aminoacids, calculating a consensus, and counting relative frequencies of aminoacids in any position of the reverse transcriptase from 71 to 294. We applied the same procedure to the RT portion of a set of full genome HBV sequences from Genbank (n=4579). Frequency analysis of these genbank sequences revealed very low prevalence of resistance-associated mutations (among those previously reported in literature), thus this population could be assumed to be treatment-naive. We run a χ-square test (with Yates' and continuity correction) adjusted for multiple comparisons (with Bonferroni). The classical resistance mutations to lamivudine (M204I, M204V and L180M) were more strongly present in the population of patients under treatment at INMI (p<0.00001). Aminoacid residues involved in genotype specificity were deleted from the analysis in order to correct potential biases due to the different geographical origin of the two subset of sequences. Table 2 shows the subset mutations, among those listed in Table 1, which showed statistically significant enrichment in the treated patients population. The cut-off for the statistical significance for this analyses was set to p<0.025. These mutations were positively selected in patients by the antiviral therapies selective drive, therefore they could contribute to determine a resistance or cross-resistance to the compounds used. Given the relative frequencies of their use in the population studied, the contribution to resistance should be maximal against lamivudine, but may play a role also against adefovir, tenofovir and entecavir.

Table 2

Mutation Prevalence in 336 sequences Prevalence in 4579 P from patients under treatment sequences from GenBank

(%) (%) rt:L_115_V; 5.7 1.2 0.0004 rt:F_122_S; 1.8 0.4 0.0000 rt:N_123_T; 3.9 0.2 0.0000 rt:T_128_I; 2.4 0.5 0.0000 rt:H_135_T; 2.7 0.2 0.0000 rt:V_142_E; 3.3 0.4 0.0000 rt:F_151_L; 2.7 1.1 0.0137 rt:S_159_T; 0.9 0.2 0.0039 rt:T_184_S; 3.3 0.5 0.0000 rt:I_187_L; 2.4 0.8 0.0028 rt:A_200_G; 0.1 0.0 0.0238 rt:K_212_N; 0.6 0.0 0.0007 rt:V_214_A; 3 0.5 0.0000 rt:L_229_M; 2.1 0.3 0.0000 rt:L_229_W; 1.2 0.3 0.0043 rt:Y_245_H; 1.2 0.5 0.0210 rt:S_246_T; 1.2 0.5 0.0208 rt:Y_257_H; 4 0.4 0.0000 rt:I_266_T; 1.5 0.9 0.0006 rt:I_290_N; 1.5 0.0 0.0000 rt:V_291_G; 0.6 0.0 0.0044 rt:L 294 F 2.5 0.0 0.0000 The automated method according to the invention comprises the following steps:

(i) carrying out the genome (or a portion thereof) sequencing of the HBV present in the biological samples from the patients; downloading the mutation data of the HBV genes which are in public or proprietary databases (where a reference population is needed); memorizing the data thus obtained;

(ii) performing the input of the information and of the data generated by the sequencing, and transformation thereof into standard electronic formats for the analysis (be it FASTA, GenBank etc.) (processing); reading and memorization thereof by a parsing process; alignment (nucleotide level, then translated) between them and with the sequences of the databases present in the memory, of the sequences obtained from patients; error correction (both by ultra-deep pyrosequencing and by the Sanger method); phylogenetic analysis (the phylogenetic analysis by large sequences groups, such as those coming from pyrosequencing, is preferably carried out by transferring the data to an elaboration centre equipped with multiprocessor devices for running parallel algorithms); variation analysis of the sequences generated in step (i) to obtain prevalence and entropy data (with confidence intervals) regarding the observed variations;

(iii) translating the nucleotide sequences obtained in step (ii) into the corresponding aminoacid sequences; further correction of the errors occurred in said aminoacid sequences; analyzing the variation of the entropy and prevalence data with reference to the sensitivity to a drug or for a new phylogenetic analysis. The evaluation of the nucleotide and aminoacid variability is again carried out with calculation of the confidence intervals of the mutation frequencies, also with the inclusion of mutations having a frequency above 1%;

(iv) extraction of the mutations and/or of the mutation profiles generated, which are associated with drug resistance or are due to a selection pushed by drug therapy, provided that statistic univariate or multivariate tests (as reported in the previous paragraphs) are carried out (such as the phylogenetic analysis and definition of clusters).

More particularly, the method of the invention is based on the following procedures:

A. Nucleotide alignment

a. Input: a consensus sequence and a set of sequences generated by GSFLX (or a set of sequences coming from a Sanger sequencing, in the latter case the statistical error correction is not carried out). i. The consensus sequence is defined by the end user, or either, is generated by a whole genome assembly program if the sequences correspond to a re-sequencing of the "shotgun" or "amplicon" type (such as CAP3, described in "PCAP: A Whole-Genome Assembly" Program. Genome Research, 13: 2164-2170.) ii. The sequences coming from the GSFLX are aligned one by one to the consensus by using a modified version of the Smith- Waterman- Gotoh local alignment algorithm (Smith and Waterman, 1981; Gotoh, 1982), possibly with optimization of parameters and choice of the scores iii. All the nucleotide variations, including insertion-deletion ones, are statistically analyzed, and possible errors in the pyrosequencing are deleted (as described in Eriksson N, Pachter L, Mitsuya Y, Rhee S- Y, Wang C, et al. (2008) "Viral Population Estimation Using Pyrosequencing. " PLoS Comput Biol 4(5): el000074. doi: 10.1371/ journal. pcbi.1000074 e in Wang C, Mitsuya Y, Gharizadeh

B, Ronaghi M, Shafer RW. "Characterization of mutation spectra with ultra-deep pyrosequencing: application to HIV-I drug resistance. " Genome Res. (2007) 17(8):1195-201.) iv. Prevalence and entropy of the nucleotide variations are calculated v. Optionally, a new consensus is calculated.

b. Note: in the case of the sequencing on multiple or very divergent species (thus NOT in the case of such viruses as HBV/HCV/HIV or similar) a multiple alignment run in parallel instead of the local alignment can be used (as described in "ClustalW-MPI: ClustalW Analysis Using Distributed and Parallel Computing", Kuo-Bin Li, Bioinformatics, (2003), 19(12), 1585-1586).

c. Output: correct nucleotide alignment i. Nucleotide variations graphs ii. A new file with aligned and correct sequences is generated iii. In the case of ultra-deep sequencing limited to small portions of the genome (so that they can be covered by the length of the generated fragments) or either in case of sequences obtained by the Sanger method, the generated files are ready for the parallel phylogenetic analysis (A. Stamatakis. "RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. " Bioinformatics,. 22(21), 2688-2690).

B. Translation of the aligned nucleotide sequences into aminoacids (i.e., polymerase and surface antigene in HBV)

a. Further correction on frameshifts and insertions which are not in frame

b. Calculation and generation of graphs of the aminoacid prevalence and entropy i. If experiments regarding the evaluation of resistant mutant selection pushed by drug therapy are being carried out, the mutated positions potentially associated with selection/resistance are reported, using different populations (e.g., treated/naive), by adjusted statistical univariate confrontations (proportion test, x square test, Bonferroni or Benjamini-Hochberg adjustments).

c. Identification of transmission/selection cluster (multivariate analysis) by phylogenesis (as per item A.c.iii). C. Quasi-species reconstruction

a. If sequencing was carried out on genome portions longer than the sequences (i.e., "shotgun" or "amplicon" sequencing), programs for the reconstruction of the quasi-species (ShoRAH, http://www.bsse.ethz.ch/cbg/software/shorah) are used.

b. The reconstructed quasi-species are ready for the phylogenetic analysis as per item A.c.iii.

To process the great number of data obtained by pyrosequencing the following procedures were performed. A series of programs in open source language Java (http://java.sun.com), which can be run on any platform type (i.e., Windows, Mac, Linux, Unix, Solaris) work as a wrapper by incorporating an implementation of SWG algorithm, supplied with the package Jaligner (jaligner.sourceforge.net).

The wrapper also includes different input-processing functions, starting from the elaboration of multiple sequences already introduced, to the re-formatting of fasta standard starting form different formats, input for the phylogenetic programs (by using the program Java readseq, http://iubio.bio.indiana.edu/soft/molbio/readseq/iava/). It is also possible to process directly the output files of the pyrosequence devices by connecting directly to a central server. Subsequently, on the basis of the alignments obtained by running the SWG algorithm, the different analysis, aminoacid translation, and output formatting procedures are run. The SWG algorithm is specifically optimized for the data generated by pyrosequencing (Wang et al., 2007): refinements are actually applied to the original algorithm for the frameshift detection/correction, for the re-alignment of insertions and deletions which are not in the correct frame, for the translation of nucleotide ambiguities, with the application of specific inclusion thresholds for variations according to Eriksson et al. (2008) and Wang et al. (2007), which analyzed in detail the error rate in the data coming from the pyrosequencing. Specifically, the fasta sequences from the pyro- sequencing device (which are read and processed as described above directly by the central servers) are subsequently selected according to the quality scores generated by the device and are then locally aligned using a consensus sequence chosen by the user or automatically determined by default (e.g., for HBV a library of different genotypes is gathered in GenBank), or generated by the same pyrosequencing data (by means of whole genome assembly, see item A.i.).

The alignment parameters that concern the penalties for gap opening and length are optimized through a grid search in real space [5,30] and [0.25,3] with spacing 5 and 0.5, respectively. It is also possible to optimize the alignment parameters by maximizing the similarity coefficient, the alignment score, or by minimizing the frameshift number by 1 base or 2 bases. Actually, since the correct frame of the protein code of the consensus is known, it is possible to divide in codifying triplets the aligned nucleotides and to operate a series of refinements on the alignment and quality checks. When insertions or deletion of three (or multiples of three) bases in a row do not correspond exactly to an in-frame triplet, an automatic procedure corrects the alignment according to a criterion of maximum plausibility and it re-positions the insertions or deletions in frame. In case of 1- or 2-base frameshifts with respect to the consensus, on the contrary; these are considered as possible errors of the pyrosequencing device and are ignored, while for 1- or 2- base frameshifts in a query sequence, the same are reported and analyzed for their frequency.

Other wrappers for the are implemented for the execution of the BLAST program (Altschul et al., 1990): specifically, Java scripts carry out BLAST and do the automatic parsing of the output using local replicas on genomic databases servers downloaded and periodically updated from the worldwide databases (GenBank, www.ncbi.org): up to now the wrapper is configured on HIV, HBV, HCV and Herpes.

Serially to the implementation of the nucleotide alignment with optimization, of the correction/detection of the errors and to the aminoacid translation, a program works to extract the nucleotide and aminoacid mutations from any consensus sequence, either on an input from a pyrosequencing device and on Sanger sequences.

The prevalence of the mutations is calculated as relative frequency in the sequenced (sub)regions, with a robust estimate of the confidence intervals and other basic descriptive statistics together with the statistical correction of errors, both at a nucleotide and aminoacid level.

Additional procedures format the alignment outputs in fasta or phylyp format, report the translated protein sequences, binary matrices for the analysis of point variation or real matrices for the quantitative analysis of the population.

These outputs can be directly used by other bioinformatic packages, such as EMBOSS (http/Zemboss. sourceforge.net), R (http/Zwww.r- proiect.org) and phylyp (http://evolution.genetics.washington.edu/phylvp.html). RaxML (phylobench.vital-it.ch/raxml-bb/), PHYML

(http://atgc.lirmm.fr/phyml/), also replicated on local devices, which are widely used by researchers.

Many thousands of sequences may thus be processed and made available for high level analysis within two hours, owing to the procedure parallelization, whereas processing on normal desktop computers implies time periods in the order of days or weeks (and the phylogenetic analysis for sequences with cardinalities above one thousand is practically impossible)

In addition, further scripts allow the execution of the bioinformatic programs mentioned above by supercomputers and by local high storage capacity servers, as a superior elaborative power is necessary for the large quantity of data coming from the pyrosequencing. It could, for example, be used the server used at the Consorzio Inter- universitario per Ie Applicazioni di Supercalcolo per l'Universita e Ia Ricerca (CASPUR, www.caspur.it).

As far as the the reconstruction of complete genomes by pyro- sequencing, a wrapper was implemented which uses the CAP3 algorithm (Huang and Madan, 1999): CAP3 is directly executed on the pyrosequencing device and it then possibly sends the generated contiguous to the above mentioned BLAST wrapper. Finally, as far as the aplotype reconstruction is concerned, this is done by implementing a work done by Eriksson et al (2008) which reassembles the most likely subpopulations from a series of data coming fron the pyrosequencing, also associating the prevalence of the single species.

The automated method of the invention allows manipulating a large number of sequences, tipically in the range of tens of thousands, either coming from Sanger sequencing and from ultra-deep pyrosequencing. Said method of the invention allows the reduction of processing timing and of the manual intervention, the execution on dedicated servers and on supercomputers. The automated method of the invention does the nucleotide alignment, the aminoacid translation, the mutation extraction, the calculation of the frequencies with confidence intervals, the reassembly of aplotypes, the subtyping, the hierarchic clustering, the generation of bootstrap samples for population studies. In addition, said method proceeds to identify mutations in the sequences obtained (up to 20,000 readings for each patient) by means of confrontation with the databases available for consultation (e.g., GenBank, the Japanese database (s2asO2.genes.nig.ac.jp), etc.). The method allows the identification of mutations also in subpopulations which would not be detected by phenotyping tests because of their presence in very poorly represented variants in the samples.

Figure 1 is a block-scheme of an exemplary operative implementation of the computerized method of the invention. Said method is advantageously implemented by means of a software which encodes the different steps of the method in a machine language when this program is run on a computer. Therefore, either said software and the computerized reading means on which the software is recorded are comprised in the scope of protection of the invention.

The computerized method of the invention also allows to carry out the genomic analysis for the prediction of the mutant viral isolates resistance to single drugs and to their combinations, by means of machine learning techniques, with statistic validation.

In practice, the method of the invention allows to input on computer, in one single session, a number of sequences obtained either by classic sequencing (typically, according to Sanger) or by ultra-deep pyrosequencing, to align them, to extract the mutation present, to make a confrontation with known mutations in GenBank or in other databases, to calculate their frequencies with their relative confidence intervals, and to predict the resistance of the viral population isolated from a single patient to one or more antiviral drug.

It is so possible to input the data from the sequencing of single patients obtained by ultra-deep pyrosequencing, or either a very large number of data obtained by classic sequencing (e.g., Sanger), and to analyze them in a short time (about 2 hours), whereas, by using the methods currently known in the art, it would be necessary to manually carry out each single step, taking weeks or months for the interpretation of the sequencing data to be carried out.

The automated method of the invention brings to an advancement in the state of the art with respect to the pre-existing automated methods, such as the built-in softwares furnished with the interface servers of the pyrosequencing devices, specifically GSBrowser, GSAmplicon, GSAssembler, GSMapper by Roche for the 454 device (http://www.454.com).

Actually, a local alignment algorithm is implemented which is optimized with detection/correction procedures of errors and automatic translation to proteins with correct frames. In addition, a series of descriptive statistics and variation analysis is produced not only at the nucleotide level, but also at the aminoacid level, with a robust estimate of the confidence intervals, a feature that is not found (or is present only for nucleotide alignments and implemented with different techniques) in the above mentioned softwares.

This flexibility in the switch from the nucleotide analysis to the aminoacid one allows the detection of sets of aminoacid substitutions of interest for the resistance analysis. Moreover, the method envisages the possibility to organize in series the different procedures of alignment, translation, extraction of mutations, variability matrices generation, of descriptive statistics, all managed by dedicated scripts which generates output files for the immediate use, while the default softwares which come with the above mentioned devices can be used only by highly specialized staff and envisage the esclusive installation on dedicated servers with particular operative systems (Linux only), with a number of limited formats for other types of analysis (currently, only ace and fasta formats).

The method uses the advanced scripts to generate a quantity of data sufficient for the investigation of mathematical methods (statistics and machine learning) to correlate detected mutations (or mutation patterns) to drug treatment resistance (or drug treatment combinations), using at the same time demographic and/or viral/immunologic data coming from clinical centres and envisaging a check of the significant univariate associations (or of linear or nonlinear multivariate models tested) both with a validation on said in vivo data and with a validation on phenotypization tests in vitro.

The following examples are meant to illustrate the invention and in no way are to be considered as limiting its scope.

Examples

Patients Blood samples from 13 patients with HBV were tested, of which 9 with genotype D and 4 with genotype A; 5 were naϊve and 8 were being treated with antivirals (4 with lamivudine, 1 with emtricitabine + tenofovir, and 3 with adefovir after a previous treatment with lamivudine).

All these patients had the virus in their blood; the HBV DNA values were obtained with the COBAS Taq-Man HBV test (Roche Molecular Systems, Inc., Branchburg, NJ, detection limit: 12 IU/ml). In particular, the values for the viral loads of treated patients were as follows: pt. No. 1: 971.104 UI/ml; pt No. 2: 1.609.456 UI/ml; pt No. 3: 40.004 UI/ml; pt No. 4: 89.700.000 Log UI/ml; pt No. 5: 530 UI/ml; pt No. 6 and No. 8: >110.000.000 UI/ml; pt No. 7: 73.659.072 UI/ml; for the naϊve patients: pt No. 9 and No. 11: >110.000.000 UI/ml; pt No. 10: 698.962 UI/ml; pt No. 12: 13.129 UI/ml; pt No. 13: 48.830.532 UI/ml.

Ultra-deep pyrosequencing was carried out in sera from these 13 patients to identify mutations in rt and in HBsAg; at the same time in all the patients the classic test of genotypic resistance was done, which is based on the direct sequencing of the same region, and in the 8 treated patients the DRv2 test was also carried out.

For all the genotypic resistance analyses, the HBV DNA was extracted from the sera with the QIAmp DNA Blood Mini kit (Valencia, CA).

Analysis by ultra-deep pyrosequencing to identify mutations in rt and in HBs

The primers (see Table 3) were designed so that they could bind to conserved sequences among the different genotypes and to amplify 8 partially overlapping segments as this technique does not allow to sequence DNA fragments longer than 250 bp (base pairs). In addition to the sequence described in the Table, the primers for ultra-deep pyrosequencing contain at the 5' end the adaptor sequences indicated by the GS FLX device, needed for the binding to the beads (left GCCTCCCTCGCGCCATCAG., right GCCTTGCCAGCCCGCTCAG) not shown in the Table.

These amplicons cover the rt gene from aminoacid 1 to aminoacid 288 (a segment which includes all the functional domains) and the complete region of the gene that codifies for HBsAg.

The conditions used for the PCR were as follows: one cycle at 95°C for 2 min followed by 40 denaturing cycles of 30 sec at 95°C, pairing of primers for 30 sec at 60⁰C and extension for 45 sec at 72°C, and finally a further step of 5 min for the extension.

Since the quality and quantity of the DNA samples are important for the success of the procedure, the 8 amplicons obtained for each sample were purified by means of a purification kit, QIA quick PCR (Qiagen, Chatsworth, CA, USA) and subsequently quantified by the Agilent 2100 bioanalyzer (Agilent Life Sciences and Chemical Analyses, CA, USA). The purified amplicons were then mixed in equal parts and underwent an ultra-deep pyrosequencing using the platform GS-FLX (Roche). A frequency of 1% was chosen as a threshold for the mutations.

The principle of such technique is summarized as follows: the DNA fragments are amplified using pairs of primers that include specific sequences called "A and B adaptors". Subsequently, the DNA fragments are denatured and mixed with the beads which have complementary oligonucleotides for the adaptors on their surface. This step is carried out with very low concentrations of DNA to obtain an average of a single DNA chain bound to one bead. The DNA bound to the beads is then amplified in a water/oil emulsion where each different drop contains one single bead with one single DNA fragment bound. At the end of the process, beads are obtained with many copies of homogeneous PCR products. The beads with the bound DNA are then distributed on a microplate at a density of about 400,000 beads/plate which is placed inside the device where the sequencing will take place, i.e. the GS FLX. A polymerase is used to elongate the DNA chain starting from the primer bound on each strand. The 4 nucleotide triphosphates are sequentially added to the microplate. Each time a nucleotide is incorporated, pyrophosphate is released (hence "pyrosequencing") which, thanks to an enzyme, is incorporated in ATP. In turn, ATP activates a luciferase which produces a light signal for each well, which is quantified, and the correspondent signals are detected and stored in the computer. The sequential application of the 4 nucleotides allows the formation of DNA fragments as long as about 260 nucleotides, obtaining in the whole about 100 million bases in a few days. All the images are then processed by the dedicated software.

The data thus obtained were processed according to the computerized method of the invention described above.

Table 3

Sequence and position of primers used for the ultra-deep pyrosequencing of HBV. In addition to the sequence described in this Table, the promers contain at the 5' end the adaptor sequences indicated by the GS FLX system, needed for linking to the beads (SEQ ID No 17 left GCCTCCCTCGCGCCATCAG, SEQ ID No 18 right GCCTTGCCAGCCCGCTCAG), not shown in the Table.

SEQ ID No Name Sequence (5'-3') Position in Rt

1 HBV 1 FW CCTGCTGGTGGCTCCAGTT -34 -53

2 HBV 1 RW AGAGAAGTCCACCACGAG 124 - 140

3 HBV 2 FW CCTGCTCGTGTTACAGGCG 58 - 72

4 HBV 2 RW CCGCAGACACATCCAGCG 242 - 259

5 HBV 3 FW CCGTGTGTCTTGGCCAAA 162 - 179

6 HBV 3 RW GACAAACGGGCAACATAC 330 - 347

7 HBV 4 FW GCTGCTATGCCTCATCTTCT 286 - 305

8 HBV 4 RW GAYGATGGGATGGGAATAC 471 - 489

9 HBV 5 FW GCACGACTCCTGCTCAAGG 396 - 414

10 HBV 5 RW CCCKACGAACCACTGAACAA 561 - 580 11 HBV 6 FW GTATTCCCATCCCATCRTC 471 - 489

12 HBV 6 RW CGGTAWAAAGGGACTCAMG 649 - 667

13 HBV 7 FW TTGTTCAGTGGTTCGTMGGG 561 - 580

14 HBV 7 RW GGGTTAAATGTATACCCAVAG 689 - 710

15 HBV 8 FW CKTGAGTCCCTTTWTACCG 649 - 667

16 HBV 8 RW CKTGAGTCCCTTTWTACCG 649 - 667

Direct sequencing and DRv2

The PCR conditions for direct sequencing were as follows: 2 groups of primers (HBVl fw - HBV4 rw and HBV5 fw - HBV8 rw) to amplify a segment having the same length (1-288 aminoacids) as the one sequenced by GS FLX (Table 3). The PCR conditions reported above. The amplified PCR product was purified by using the QIAquick PCR purification kit (Qiagen, Chatsworth, CA, USA) and quantified by gel electrophoresis with a known molecular weight standard. The sequencing was carried out by means of the ABI Prism 3100, using the BigDye Terminator Cycle Sequencing (Applied Biosystems, Warrington, UK).

In the samples from patients undergoing anti-HBV treatment, the resistance test was carried out also by INNO-LiPA HBV DRv2 (Innogenetics NV, Ghent, Belgium) following the fabricant's instructions.

Sequences representative of the A and D genotypes of HBV from GeneBank were used (genotype A: AAY25236, AAY25270, EOOOlO, V00866; genotype D: Y07587, AF043594, Y796031, AY721610, AY796030), on which the consensus sequences were obtained.

Results and conclusions

290,315 reads were obtained in the whole, with an average length of 189 bp; the range of the reads number for each patient was 2,852 - 180,167. In Table 5 the mutations found with the GS FLX in 8 patients undergoing treatment are shown, in comparison with those identified by direct sequencing and by DRv2. All the mutations at a frequence <24% by GS FLX and DRv2 were not identified by direct sequencing. In 3 patients some mutations (4 for patient 1, 3 for patient 5 and 1 for patient 6) were detected by GS FLX at a frequency in between 2% and 8% and went undetected by DRv2. Only in patient 4 a nutation (M204V) was detected by DRv2, but not by GS FLX and by the direct sequencing (the DRv2 result could result from an artefact).

These results show that all the most important mutations associated with resistance which are included in DRv2 were identified by GS FLX. In addition, the GS FLX device detected further mutations (not comprised in DRv2) at a variable sequence (>1%) both in treated and in naϊve patients. The listing of all the mutations found is reported in Table 6.

It is be pointed out that in 2 naϊve patients GS FLX was able to identify mutations associated with resistance, in particular V214A, Q215S, M204I with a frequency of 6%, 26%, 1% respectively, for patient 11, and V214A, Q215H with a frequency of 2%, 1% respectively, for patient 12. Anyway, the primary mutation M204I, identified with a low frequency in patient 11 corresponds to a stop codon in HBsAg.

The total distribution of all the mutated codons identified by GS FLX in the rt gene in each patient is shown in Figure 2, where the y-axis has been cut at 20% (direct sequence lower limit) to better visualize the less frequent mutations.

In this Figure it can be seen that, as known in the literature, the D genotype is more eterogeneous than the A genotype both in treated patients and in naϊve ones. In addition, both in treated patients and in naϊve ones mutated positions can be found, both in functional domains and between domains, the total number of mutations was always higher in the treated patients with respect to the naive ones both in the functional domains (48 vs. 19) and between domains (96 vs. 63). Anyway, the ratio between treated patients and naϊve patients was higher in functional domains than in between domains (2.5 vs. 1.5, respectively).

In treated patients the method used allowed the identification of new mutations, never described before in the literature according to our knowledge, in 45 different codons, of which 15 were also present in naϊve patients. Nine out of these mutations were present in the viral quasispecies with a frequency >24% (i.e., they could also be seen by direct sequencing), 2 mutations were found in 2 patients with a very different frequency (1 vs. 99%) and the remaining mutations had a frequency <24% in all patients.

GS FLX sequencing also allowed to identify and to quantify mutations in the HBV surface antigene for each patient (Table 5). The correspondence between the mtations in rt and those in HBs is shown in Table 6.

In the whole, the described method, based on ultra-deep pyrosequencing by use of the sequencing platform GS FLX and on use of the automated method of the invention, allowed to analyze the HBV quasispecies in the region codifying for the reverse transcriptase and for HBsAg, highlighting the presence of (i) mutations that were already known for being associated with resistance to different drugs, but with the innovating feature that their frequency can be calculated in each patient; (ii) mutations present with a frequency lower than 20%, which could not be seen by classic sequencing, (iii) known mutations, present with a frequency lower than 5%, which could not be seen by use of DRv2; (iv) mutations that were never described before, whose frequency ranged in between 1 to 99%.

The quantitative feature pointed out in the described examples is important as, followed up in time in each single patient, it allows to evaluate an evolving situation and, therefore, to carry out changes in the th _IRe.rapy before the mutated variant definitely establishes itself in the viral quasispecies.

Also, the feature r _IRe. lating to the new mutations identified is important since it allows to re-evaluate the algorithms used to interpret the genotypic resistance tests.

A further important aspect is the identification of mutations in HBsAg which are not all described in the literature, and it can can applied to the study of immune escape variants. In addition, some mutations in rt which determine stop codons in HBsAg have been described in the literature, and for them a role has been pro _IRp.osed in the maintenance of resistance (thanks to the change made to the replication enzyme), but only in co-presence of the correspondent wt, which furnishes the missing HBsAg function due to truncation in the mutated strain. These cases are very rare, but it is not known if this is due to the fact that direct sequencing does not allow to see mutation frequencies below 20%. By applying the automated system herein claimed it was possible to evidentiate a similar situation, never described before, in codon rt204 of patient 11, which corresponds to a stop codon in HBsAg, which is present with a frequency of 1% and which could not be appreciated with the methods currently in use.

Table 4

HBV variants obtained by sequencing with GS FLX compared to those obtained by INNO-LiPA DRv2 and by direct sequencing in patients under anti-HBV treatment.

Codon 80 Codon 173 Codon 180

L80(wt) L80I L80V V173 (wt) V173L L180 (wt) L180M

^ Q¹ S? Qf ≠ Qf ^ Qf ^ Qf ^ O" ^ Qf

¹ x w x w x w x w

J^ C -O J _1N K COI J _1N M JJ ^ (M C UOl x JJ O (M, w C COO x J w

J (M - CO x - w

(M CO fa > A <M Ul fa > A fa > fa > > >

Ul co Ul Ul Ul M fa > fa co

O P Q O P Q O Q P O Q O O Q Q O P O P

Pt I 97 N N 100 92 8 N N * %

Pt 2 100 Y Y 100 Y Y 100 Y Y

Pt 3 72 Y Y 25 Y Y 3 Y N 100 Y Y 100 Y Y

Pt 4 97 Y Y 3 Y N 100 Y Y 100 Y Y

Pt 5 100 Y Y v 100 Y Y 97 Y Y 3 N N

Pt 6 100 Y Y D_IRE. 38 Y Y 62 Y Y 40 Y Y 60 Y Y

Pt 7 81 Y Y 1 FLX * %9 Y N 100 Y Y 23 Y N 77 Y Y

Pt 8 100 Y Y 100 Y Y 59 Y Y 41 Y Y

Codon 181 Codon 204 D_IRE. Codon N236

A181 (Wt) A181T/V* M204 (wt) M2 F *LX %04I M204V N236ι (wt) N236T

Pt 1 98 Y Y 2 N N 95 Y Y 5 N N 100 Y Y

Pt 2 97 Y Y 3 Y N 100 Y Y ^SE^Q. 85 Y Y 15 Y N

Pt 3 100 Y Y 88 Y Y 12 Y N 10 G^SX ^FL %0 Y Y

Pt 4 100 Y Y 100 Y Y N Y N 100 Y Y

Pt 5 93 Y Y 7 * N N 98 Y Y 2 N N 87 Y Y 13 Y N

Pt 6 98 Y Y 2 * N N 39 Y Y 21 Y Y 40 Y Y D^IRE.

Pt 7 100 Y Y O N N 24 Y Y 76 Y Y 100 Y Y %

Pt 8 100 Y Y 59 Y Y 41 Y Y 100 Y Y v

The frequence is reported when detection is by GSFLX SE^Q

Table 5

Mutations in rt and in HBsAg, and their frequencies (with 95% confidence interval) detected by ultra-deep pyrosequencing in 9 patients with genotype D HBV and in 4 patients with genotype A HBV. For each patient the HBV DNA concentration, the administered drug and the number of reads obtained by the GS FLX platform are reported.

CI = Confidence Interval Genotype D

Pazient 1 (4317 reads)

HBV DNA 5. 99 LoglO UI/ml

Treatment = ADV

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtG4E 0.019 0.004 0.058 rtA7D 0.172 0.120 0.240 rtR18K 0.013 0.001 0.049 rtP20S 0.019 0.004 0.058 rtT24A 0.013 0.001 0.049 rtH35Y 0.022 0.011 0.041 rtA38T 0.020 0.010 0.038 rtL80I 0.040 0.029 0.053 rtI91P 0.011 0.006 0.020 rtSlO6A 0.013 0.008 0.021 rtV112L 0.782 0.742 0.818 rtNllβS 0.079 0.057 0.108 rtR120S 0.081 0.059 0.111 rtD139S 0.011 0.004 0.026 rtY141Q 0.699 0.662 0.734 rtV142E 0.187 0.158 0.219 rtK154E 0.647 0.609 0.684 rtS159T 0.995 0.966 1.000 rtH160R 0.022 0.007 0.058 rtL180M 0.075 0.056 0.099 rtA18lT 0.023 0.014 0.039 rtM204V 0.054 0.046 0.065 rtK212R 0.156 0.141 0.172 rtF221Y 0.997 0.992 0.999 rtI233V 1.000 0.589 1.000 rtC256G 0.286 0.078 0.649

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sE2A 0.013 0.001 0.049 sN3S 0.013 0.001 0.049 sGlOR 0.013 0.001 0.049 sPHL 0.013 0.001 0.049 sL13P 0.013 0.001 0.049 sV14A 0.045 0.020 0.092 sV14G 0.038 0.016 0.083 sQ54R 0.076 0.063 0.091 sW74U 0.011 0.006 0.020 sF83L 0.011 0.006 0.020 sM103I 0.778 0.737 0.814 si HOV 0.079 0.057 0.108 sGH2A 0.081 0.059 0.111 sT131A 0.011 0.004 0.026 sM133R 0.697 0.661 0.732 sM133T 0.011 0.005 0.023 sY134N 0.187 0.158 0.219 sP135H 0.185 0.157 0.217 sI152T 0.011 0.001 0.042 si 152V 0.022 0.007 0.058 sW163U 0.017 0.009 0.031 sW172U 0.032 0.020 0.049 sL173P 0.017 0.009 0.031 sI195M 0.054 0.046 0.065 sW201U 0.012 0.008 0.018 sR204G 0.156 0.141 0.172 sR207N 0.222 0.203 0.243 sI208T 0.998 0.994 0.999 sR210K 0.241 0.221 0.261 sL213I 0.996 0.991 0.998

Pazient 2 (8876 reads)

HBV DNA 6.21 LoglO UI/ml

Treatment = ADV Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtH13L 0.994 0.989 0.997 rtR18S 0.028 0.022 0.036 rtV27F 0.013 0.009 0.019 rtA38P 0.015 0.011 0.020 rtA38S 0.015 0.012 0.020 rtA38T 0.734 0.719 0.748 rtG52E 0.011 0.006 0.018 rtH54D 0.062 0.050 0.076 rtW58U 0.010 0.006 0.017 rtL115V 0.011 0.004 0.024 rtI121S 0.125 0.100 0.156 rtT128I 0.686 0.646 0.724 rtV142E 0.105 0.087 0.126 rtA18lT 0.033 0.026 0.043 rtL199V 0.011 0.008 0.015 rtF221Y 0.018 0.012 0.026 rtL229W 0.022 0.000 0.129 rtS230C 0.022 0.000 0.129 rtI233T 0.022 0.000 0.129 rtI233V 0.022 0.000 0.129 rtL235U 0.022 0.000 0.126 rtL235V 0.022 0.000 0.126 rtN236T 0.217 0.121 0.359 rtA238H 0.022 0.000 0.129 rtK24lR 0.022 0.000 0.129 rtG255R 0.022 0.000 0.129 rtE271A 0.022 0.000 0.129 rtE271D 0.044 0.005 0.159 rtR274S 0.022 0.000 0.129

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sT5S 0.996 0.992 0.998 sGlOA 0.028 0.022 0.036 sL2lS 0.148 0.137 0.161 sR24K 0.033 0.028 0.040 sW36U 0.017 0.011 0.026 sL39P 0.029 0.021 0.040 sT57I 0.022 0.017 0.028 sP70L 0.028 0.023 0.036 sS113A 0.125 0.100 0.156 sS114T 0.011 0.004 0.024 sP120S 0.685 0.645 0.722 sY134N 0.104 0.086 0.125 sCl49Y 0.012 0.004 0.030 sP151L 0.071 0.050 0.101 sA166V 0.013 0.008 0.019 sW172U 0.036 0.028 0.046 sV180A 0.010 0.006 0.017 sW182U 0.011 0.007 0.019 sR204K 0.011 0.008 0.016 sR207N 0.568 0.543 0.593 sL213I 0.018 0.012 0.026 sC22lG 0.022 0.000 0.129 sL222V 0.022 0.000 0.129 sY225H 0.023 0.000 0.131

Patient 3 (15483 reads)

HBV DNA 4.60 LoglO UI/ml

Treatment = LAM

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtV27F 0.018 0.013 0.025 rtL80I 0.243 0.226 0.260 rtL80V 0.030 0.024 0.037 rtL82M 0.034 0.027 0.042 rtM204I 0.119 0.111 0.128

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sE164V 0.556 0.533 0.577 sW196L 0.116 0.108 0.125

Patient > i (15623 reads)

HBV DNA 6.91LoglO UI/ml

Treatment = LAM

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtV27F 0.024 0.019 0.030 rtA38E 0.377 0.364 0.390 rtG52E 0.016 0.012 0.022 rtN53D 0.095 0.084 0.107 rtL80V 0.030 0.024 0.037 rtI91F 0.016 0.012 0.022 rtI91V 0.016 0.012 0.022 rtA97T 0.011 0.007 0.015 rtG104S 0.011 0.008 0.014 rtG107R 0.012 0.009 0.015 rtR114H 0.010 0.006 0.017 rtR120K 0.011 0.007 0.018 rtT128I 0.845 0.826 0.862 rtD134E 0.101 0.087 0.117 rtH135T 0.825 0.805 0.843 rtC136Y 0.010 0.006 0.016 rtR138K 0.012 0.008 0.019 rtD139K 0.029 0.022 0.038 rtF221Y 0.055 0.047 0.065 rtI266V 0.991 0.973 0.998 rtQ267H 0.011 0.003 0.031 rtE271H 0.991 0.973 0.998 rtR274G 0.011 0.003 0.031

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sR24K 0.016 0.013 0.020 sQ30K 0.384 0.371 0.397 sW36U 0.015 0.011 0.021 sE44R 0.014 0.010 0.020 sW74U 0.012 0.009 0.017 sI82M 0.016 0.012 0.022 sV106I 0.010 0.006 0.017 sP120S 0.843 0.824 0.860 sR122Q 0.017 0.012 0.025 sT126S 0.102 0.088 0.118 sA128T 0.010 0.006 0.016 sT131N 0.029 0.022 0.038 sS193L 0.052 0.045 0.059 sR207N 0.053 0.045 0.062 sL213I 0.055 0.047 0.065 sL216U 0.010 ^■ 0.007 0.015 sV224A 0.040 0.024 0.067

Patient 5 (12287 reads)

HBV DNA 2.72 LoglO UI/ml

Treatment = ADV

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtN53D 0.024 0.018 0.031 rtN53K 0.024 0.019 0.032 rtV56M 0.140 0.127 0.156 rtL80I 0.020 0.015 0.025 rtQ125H 0.011 0.006 0.019 rtK149H 0.044 0.035 0.055 rtR153W 0.128 0.113 0.145 rtK154N 0.016 0.011 0.023 rtL157P 0.014 0.007 0.028 rtL164M 0.600 0.559 0.639 rtL180M 0.033 0.026 0.043 rtA181V 0.074 0.063 0.087 rtT184S 0.030 0.023 0.039 rtA200V 0.128 0.118 0.140 rtM204V 0.023 0.018 0.028 rtV214A 0.017 0.013 0.021 rtT225S 0.428 0.409 0.448 rtL228P 0.011 0.007 0.016 rtL229S 0.012 0.000 0.075 rtS230P 0.025 0.002 0.092 rtL231S 0.025 0.002 0.092 rtI233M 0.025 0.002 0.092 rtN236T 0.123 0.067 0.216 rtP237T 0.531 0.423 0.636 rtA238H 0.025 0.002 0.092 rtA238S 0.012 0.000 0.075 rtR242G 0.012 0.000 0.075 rtG244R 0.012 0.000 0.075 rtH248Y 0.012 0.000 0.075 rtM250T 0.012 0.000 0.075 rtY252H 0.012 0.000 0.075 rtC256G 0.358 0.262 0.468 rtY257W 0.531 0.423 0.636 rtS259L 0.012 0.000 0.075 rtQ262U 0.012 0.000 0.075 rtH264Y 0.012 0.000 0.075 rtI266V 0.469 0.364 0.577 rtQ267H 0.506 0.399 0.613 rtQ267R 0.012 0.000 0.075 rtQ267Y 0.012 0.000 0.075 rtI269T 0.012 0.000 0.075 rtE271D 0.519 0.411 0.624 rtE271H 0.111 0.058 0.201

Mutations in HBsAg

Mutation Prevalence 1 L.ower 95 CI Upper 95CI sV14A 0.034 0.026 0.045 sT45N 0.024 0.018 0.031 sS55F 0.057 0.051 0.064 SS117T 0.011 0.006 0.019 sK141T 0.044 0.035 0.055 sN146T 0.016 0.011 0.023 sC149R 0.014 0.007 0.028 sG159A 0.033 0.026 0.042 sF161Y 0.032 0.025 0.042 sW163U 0.010 0.007 0.016 sA168V 0.035 0.027 0.044 sL173F 0.074 0.063 0.087 sL175F 0.030 0.023 0.039 sV177A 0.036 0.028 0.045 sT189I 0.466 0.449 0.482 sV190A 0.275 0.261 0.290 sL192F 0.128 0.118 0.140 sS193L 0.334 0.319 0.350 sV194A 0.023 0.018 0.028 sI195M 0.023 0.018 0.028 sI195T 0.022 0.017 0.027 sY206H 0.015 0.012 0.020 sI208T 0.709 0.690 0.727 sR210N 0.951 0.941 0.959 sL216F 0.425 0.406 0.445 sF220L 0.011 0.008 0.017 sC221R 0.012 0.000 0.075 sL222P 0.025 0.002 0.092 sW223R 0.025 0.002 0.092 sV224A 0.062 0.024 0.141 sY225C 0.025 0.002 0.092 sY225S 0.062 0.024 0.141

Patient 6 (4459 reads)

HBV DNA >8.04 LoglO UI/ml

Treatment = EMC+TDV

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtD2G 0.012 0.000 0.074 rtW3R 0.011 0.000 0.069 rtG4R 0.011 0.000 0.069 rtA7T 0.420 0.322 0.525 rtEllK 0.011 0.000 0.069 rtH13R 0.057 0.022 0.131 rtR22C 0.011 0.000 0.069 rtV23A 0.023 0.002 0.085 rtT24A 0.011 0.000 0.069 rtG26R 0.011 0.000 0.069 rtV27F 0.042 0.013 0.107 rtA38T 0.065 0.050 0.084 rtN53S 0.055 0.040 0.074 rtH54S 0.023 0.014 0.038 rtH54T 0.025 0.015 0.039 rtR55H 0.025 0.015 0.039 rtG107R 0.019 0.013 0.026 rtS109P 0.012 0.006 0.024 rtV112A 0.014 0.007 0.026 rtN118T 0.014 0.007 0.026 rtI121N 0.011 0.005 0.022 rtF122I 0.024 0.014 0.038 rtN123D 0.924 0.902 0.941 rtH126Y 0.029 0.019 0.044 rtD139Q 0.011 0.005 0.022 rtL145M 0.050 0.037 0.066 rtY148F 0.011 0.006 0.021 rtF151Y 0.039 0.028 0.055 rtR153W 0.012 0.007 0.022 rtL164M 0.098 0.061 0.155 rtK168E 0.012 0.003 0.037 rtG172U 0.028 0.013 0.059 rtV173L 0.598 0.536 0.658 rtL180M 0.606 0.544 0.665 rtA181V 0.016 0.005 0.043 rtA186T 0.012 0.000 0.071 rtI187T 0.012 0.000 0.071 rtI187V 0.395 0.298 0.502 rtC188R 0.012 0.000 0.071 rtR192S 0.012 0.000 0.071 rtM204I 0.210 0.190 0.230 rtM204V 0.398 0.374 0.422 rtF221Y 0.073 0.061 0.087 rtT222A 0.056 0.046 0.069

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sN3S 0.011 0.000 0.069 sT5A 0.057 0.022 0.131 sV14A 0.011 0.000 0.069 sR24K 0.047 0.034 0.064 sT45A 0.051 0.037 0.070 sT46P 0.055 0.040 0.074 sV47T 0.025 0.015 0.039 sL49P 0.036 0.024 0.052 sT57I 0.041 0.033 0.051 sP62L 0.011 0.007 0.017 sT68I 0.055 0.046 0.067 sC76S 0.299 0.273 0.326 sC76Y 0.036 0.027 0.049 sF83S 0.909 0.891 0.924 sL84P 0.018 0.012 0.028 sF85C 0.012 0.007 0.021 si HOL 0.014 0.007 0.026 sS113T 0.011 0.005 0.022 sR122K 0.022 0.013 0.036 sQ129H 0.029 0.019 0.044 sT131N 0.011 0.005 0.022 sY134F 0.056 0.043 0.074 sI140S 0.012 0.007 0.022 sS143T 0.039 0.028 0.055 sG159A 0.293 0.240 0.353 sF161Y 0.302 0.248 0.363 sW163C 0.028 0.013 0.059 sW163U 0.012 0.003 0.037 sE164D 0.594 0.532 0.654 sA168V 0.301 0.247 0.361 sF170S 0.012 0.003 0.037 sL173F 0.016 0.005 0.043 sL173P 0.012 0.003 0.037 sL175S 0.502 0.440 0.564 sF179L 0.012 0.000 0.071 sV180A 0.012 0.000 0.071 sF183L 0.012 0.000 0.071 sI195M 0.395 0.372 0.420 sW196S 0.208 0.189 0.229 sW201U 0.026 0.019 0.035 sR207N 0.076 0.064 0.091 sL213I 0.018 0.012 0.026 sL213M 0.055 0.045 0.068

Patient 11 (9329 reads)

HBV DNA >8.04 LoglO UI/ml naive

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtA7T 0.012 0.008 0.019 rtE8K 0.014 0.009 0.021 rtGlOR 0.011 0.007 0.017 rtR15K 0.017 0.012 0.025 rtR18K 0.075 0.063 0.090 rtV27F 0.034 0.026 0.043 rtA38P 0.019 0.014 0.025 rtA38S 0.019 0.015 0.025 rtA38T 0.066 0.058 0.076 rtN53D 0.082 0.068 0.099 rtS78C 0.306 0.286 0.326 rtA87E 0.013 0.007 0.023 rtA87G 0.013 0.007 0.023 rtA87V 0.015 0.009 0.026 rtF88S 0.025 0.016 0.037 rtI91F 0.011 0.006 0.021 rtSlO6A 0.165 0.149 0.181 rtG107R 0.012 0.008 0.018 rtRllOG 0.251 0.227 0.277 rtV112A 0.029 0.020 0.040 rtA113G 0.015 0.009 0.024 rtR114H 0.013 0.008 0.022 rtL115V 0.242 0.218 0.268 rtN118Y 0.013 0.008 0.022 rtR120K 0.011 0.007 0.020 rtI121S 0.011 0.007 0.020 rtI121V 0.022 0.015 0.033 rtF122S 0.014 0.009 0.023 rtN123T 0.017 0.011 0.026 rtH126Y 0.979 0.969 0.986 rtM129L 0.010 0.005 0.018 rtD13lS 0.041 0.031 0.055 rtD134E 0.035 0.026 0.048 rtD134V 0.614 0.584 0.643 rtC136Y 0.010 0.006 0.019 rtR138K 0.018 0.011 0.028 rtV142I 0.031 0.025 0.039 rtFlδlV 0.019 0.014 0.025 rtR153Q 0.316 0.297 0.336 rtI163V 0.098 0.081 0.118 rtA181S 0.011 0.007 0.016 rtA200T 0.011 0.007 0.016 rtM204I 0.015 0.010 0.021 rtD205N 0.013 0.009 0.019 rtV214A 0.070 0.059 0.082 rtE218D 0.219 0.195 0.245 rtF221Y 0.740 0.713 0.767 rtL235U 0.012 0.005 0.026 rtA238H 0.018 0.009 0.034 rtY257H 0.528 0.485 0.572 rtE271D 0.025 0.014 0.044 rtI278A 0.016 0.007 0.031

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sT5I 0.064 0.053 0.077 sG7R 0.012 0.008 0.019 sF8H 0.046 0.036 0.057 sF8P 0.066 0.055 0.080 sF8R 0.017 0.011 0.024 sGlOK 0.060 0.049 0.073 sGlOR 0.016 0.011 0.023 sV14A 0.011 0.007 0.018 sV14G 0.084 0.072 0.099 sW36U 0.171 0.151 0.193 sQ51R 0.078 0.064 0.094 sS58F 0.010 0.007 0.016 sP70A 0.303 0.283 0.324 sC76F 0.030 0.020 0.044 sC76S 0.029 0.019 0.043 sC76Y 0.203 0.177 0.231 sR78Q 0.255 0.227 0.285 sR79C 0.016 0.009 0.028 sR79G 0.014 0.008 0.025 sR79S 0.015 0.009 0.026 sF80L 0.025 0.016 0.037 sF80S 0.213 0.188 0.241 sI86R 0.025 0.017 0.037 sQlOIR 0.037 0.028 0.050 sM103I 0.010 0.006 0.019 sP105A 0.017 0.011 0.027 sP105S 0.011 0.006 0.020 sV106G 0.012 0.007 0.021 sV106I 0.010 0.006 0.019 sC107R 0.017 0.011 0.026 sC107Y 0.016 0.010 0.025 sGll2E 0.011 0.006 0.019 sG112R 0.019 0.012 0.028 sS113A 0.012 0.007 0.021 sS114P 0.016 0.010 0.025 sT115P 0.017 0.011 0.026 sR122Q 0.016 0.010 0.026 sT123V 0.042 0.032 0.056 sT126S 0.645 0.615 0.673 sA128T 0.010 0.006 0.019 sG130E 0.012 0.007 0.021 sG130R 0.013 0.008 0.022 sM133I 0.031 0.025 0.040 sY134C 0.027 0.021 0.035 sC139Y 0.010 0.007 0.016 sK141U 0.017 0.013 0.024 sS143L 0.011 0.008 0.017 sG145E 0.016 0.012 0.023 sG145R 0.312 0.293 0.332 sC147Y 0.014 0.010 0.020 si 150T 0.025 0.017 0.037 sG159E 0.011 0.007 0.016 sW163U 0.012 0.008 0.018 sW172C 0.010 0.006 0.016 sW182U 0.012 0.006 0.022 sW191U 0.012 0.009 0.018 sW196U 0.020 0.015 0.027 sW199U 0.012 0.008 0.018 sR204N 0.012 0.008 0.017 sY206H 0.070 0.059 0.082 sR207C 0.039 0.029 0.052 sR207G 0.039 0.029 0.052 sI208T 0.050 0.039 0.065 sR210T 0.216 0.193 0.242 sP211L 0.100 0.084 0.120 sL213I 0.684 0.655 0.712 sL213T 0.063 0.049 0.079 sP214L 0.043 0.032 0.058 sL216U 0.058 0.047 0.071 sI218T 0.015 0.010 0.023 sF219S 0.016 0.011 0.024 sF220C 0.044 0.034 0.056 sV224A 0.025 0.015 0.044 sY225S 0.028 0.016 0.047

Patient 12 (5667 reads)

HBV DNA 4.12 LoglO UI/ml naive

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtV27F 0.052 0.038 0.071 rtF122N 0.032 0.018 0.055 rtH126Y 0.032 0.018 0.056 rtD134G 0.013 0.005 0.032 rtD139Q 0.032 0.018 0.055 rtL145M 0.018 0.011 0.028 rtF151Y 0.017 0.010 0.027 rtR153W 0.019 0.012 0.029 rtS159T 0.977 0.961 0.986 rtI163V 0.012 0.006 0.025 rtV214A 0.019 0.014 0.026 rtP215H 0.016 0.011 0.023 rtR242G 0.012 0.005 0.026

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sPHL 0.020 0.011 0.034 sT118A 0.013 0.005 0.032 sT126A 0.013 0.005 0.032 sS193L 0.064 0.054 0.075 sY206H 0.019 0.014 0.026 sN207I 0.047 0.033 0.067

Patient 13 (180167 reads)

HBV DNA 7.69 LoglO UI/ml naive

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtV27F 0.045 0.036 0.055 rtN53D 0.472 0.456 0.488 rtH54D 0.442 0.426 0.458 rtM129L 0.012 0.008 0.017 rtD134E 0.056 0.047 0.066 rtD134V 0.548 0.527 0.568 rtD139K 0.076 0.065 0.087 rtR153W 0.989 0.985 0.992 rtP237T 0.995 0.969 1.000 rtY245H 0.066 0.038 0.111 rtY257G 0.020 0.006 0.053 rtY257W 0.949 0.908 0.973 rtQ267H 0.980 0.947 0.994 rtE271D 0.934 0.889 0.962 rtR274Q 0.010 0.000 0.039

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sT126S 0.602 0.582 0.622 sT131N 0.076 0.066 0.088

SS193L 0.349 0.335 0.363 sR204N 0.013 0.010 0.017 sR207N 0.016 0.011 0.022 sI208T 0.605 0.585 0.625 Genotype A

Patient 7 (8917 reads)

HBV DNA 7.87 LoglO UI/ml

Treatme = LAM

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtH13N 0.183 0.161 0.209 rtH53S 0.190 0.167 0.216 rtL80V 0.182 0.162 0.204 rtI103V 0.039 0.030 0.051 rtV173L 0.047 0.038 0.057 rtL180M 0.772 0.752 0.790 rtM204I 0.238 0.226 0.250 rtM204V 0.754 0.741 0.767 rtI253T 0.050 0.000 0.258 rtR274S 0.048 0.000 0.248

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sT45A 0.191 0.168 0.217 sT45S 0.810 0.784 0.833 sP67Q 0.121 0.108 0.135 sPlllL 0.013 0.005 0.031 sE164D 0.046 0.037 0.057 sI195M 0.754 0.741 0.767 sW196L 0.238 0.226 0.251 sL209V 0.996 0.994 0.998

Patient 8 (10590 reads)

HBV DNA >8.04 LoglO UI/ml

Treatment = LAM

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtV27F 0.017 0.012 0.024 rtH53S 0.235 0.215 0.257 rtI103V 0.041 0.032 0.052 rtL180M 0.409 0.387 0.430 rtM204V 0.421 0.405 0.437 rtP237L 0.011 0.001 0.042 rtS246P 0.011 0.001 0.042 rtF249L 0.011 0.001 0.042 rtM250I 0.016 0.004 0.050

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sT45A 0.235 0.215 0.257 sT45S 0.764 0.743 0.784 sW172U 0.010 0.007 0.016 sI195M 0.420 0.404 0.436 sL209V 0.993 0.989 0.996

Patient 9 (11768 reads)

HBV DNA >8.04 LoglO UI/ml naive

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtV27F 0.045 0.037 0.055

Mutations in HBsAg

Mutation Prevalence Lower 95 CI Upper 95CI sT45S 0.996 0.991 0.998 sL209V 0.999 0.994 1.000

Patient 10 (2852 reads)

HBV DNA 5.84 LoglO UI/ml naive

Mutations in rt

Mutation Prevalence Lower 95 CI Upper 95CI rtW153R 0.996 0.987 0.999 rtL229V 0.015 0.005 0.035 rtP237T 0.468 0.415 0.521 rtN248H 0.462 0.410 0.515 rtS256C 0.591 0.537 0.642 rtW257L 0.120 0.089 0.159 rtT259P 0.129 0.097 0.169 rtT259S 0.462 0.410 0.515 rtQ267H 0.468 0.415 0.521 rtH271D 0.462 0.410 0.515 rtH271E 0.129 0.097 0.169 rtV278I 0.416 0.364 0.471 rtR280M 0.979 0.879 1.000 rtD283E 0.935 0.816 0.984 rtW284R 0.024 0.000 0.140

Mutations in HBsAg

Mutatzion Prevalence Lower 95 CI Upper 95CI sT45S 0.997 0.980 1.000 sYlOOC 0.994 0.984 0.998 sR207N 0.992 0.974 0.998 sF220L 0.013 0.006 0.026

Table 6.

Mutations in rt and corresponding mutations in HBsAg, determined by considering that each mutation of HBsAg can be generated by substitutions in each of the 2 adiacent codons of rt

rt mutation #1 rt mutation #2 mutations in

HBsAg rtEHK - sN3S rtH13L - sT5S rtH13R - sT5A rtR15K - sG7R rtR18K - sGlOK rtR18K - sGlOR rtR18S - sGlOA

- rtP20S sPHL rtR22C rtV23A sV14A rtR22C sV14A - rtV23A SV 14A rtG52E rtN53D sG44R rtG52E - sG44R rtN53K - sT45N rtN53S - sT45A rtN53S rtH54S sT45A rtN53S rtH54T sT45A rtH54S rtR55H sT46P rtH54S - sT46P rtH54T rtR55H sT46P rtH54T - sT46P rtR55H - sV47T rtS78C - sP70A

- rtA87E sR78Q

- rtA87G sR78Q

- rtA87V sR78Q rtA87E - sR79S rtA87E rtF88S sR79S rtA87G - sR79G rtA87G rtF88S sR79G rtA87V - sR79C rtA87V - sR79S rtA87V rtF88S sR79C rtA87V rtF88S sR79S rtF88S - sF80L

- rtL91V sI82M rtL91P - sF83L

- rtRllOG sQlOIR

- rtV112L sM103I rtA113G rtR114H sP105A rtA113G - sPlOδA rtR114H - sV106I

- rtLllδV sV106G rtL115V - sC107Y rtD118S slllOV rtD118T - si HOL rtR120K - sG112R rtR120K rtI121S sG112R rtR120K rtI121V sGH2R

- rtI12lS sG112E

- rtI12lV sG112E rtR120S - sG112A rtI121N - sS113T rtI121N rtF122I sS113T rtI121S - sS113A rtI12lS rtF122S sS113A rtF122S - sS114P rtF122S rtN123T sS114P rtN123T - sT115P rtQ125H - sS117T rtT128I - sP120S rtN13lS - sT123V rtD134E - sT126S rtD134G - sT126A rtD134V - sT126S rtC136Y - sA128T rtR138K - sG130R rtN139K - sT131N rtN139Q - sT131N rtN139S - sT131A rtY14lQ rtV142E sM133R rtY14lQ - sM133R

- rtV142E sM133T

- rtV142I sM133I rtV142E - sY134N rtY148F - sT140S rtQ149H - sK141T rtF151V - sS143L rtFlδlY - sSl43T rtR153Q SG145R rtK154N - sN146T rtL157P - sC149R rtH160R - sI152V

- rtK168E sGl59A

- rtGl72U sW163C

- rtV173L sE164D rtG172U rtV173L sE164D

- rtA181S sW172C

- rtA181T sW172U rtL180M rtA181T sW172U rtA181T - sL173P rtA181V - sL173F

- rtT184S sL175F

- rtC188R sF179L rtI187T - sF179L rtI187T rtCl88R sF179L rtI187V rtC188R sF179L rtCl88R - sV180A

- rtR192S sF183L

- rtA200T sW191U rtA200V - sL192F

- rtM204V sI195M

- rtM204V sI195T rtM204I - sW196L rtM204I - sW196S rtM204I - sW196U rtM204I rtD205N sW196U

- rtD205N sW196U rtK212R - sK204G rtV214A - sF206H rtE218D - sS210T rtF221Y rtT222A sL213M rtF221Y - sL213I rtF221Y - sL213T rtT225S sL216F - rtL229S sF220L

- rtL229V sF220L rtL228P - sF220L rtL228P rtL229S sF220L rtL229S rtS230P sC221R rtL229S - sC221R rtL229W rtS230C sC22lG rtL229W - sC221G rtS230C - sL222V rtS230P - sL222P rtS230P rtL23lS sL222P

- rtL231S sL222P rtL231S - sW223R

- rtI233M sV224A rtI233M - sY225C rtI233T sY225H

References

Peters M, et al. Walling DM, Kelly K, Davis GL, Waggoner JG, Hoofnagle JH. Immunologic effects of interferon-alpha in man: treatment with human recombinant interferon-alpha suppresses in vitro immunoglobulin production in patients with chronic type B hepatitis. J Immunol. 1986 Nov 15;137(10):3147-52.

Marcellin P, et al. Lau GK, Bonino F, Farci P, Hadziyannis S, Jin R, Lu ZM, et al. Peginterferon Alfa-2a HBeAg- Negative Chronic Hepatitis B Study Group. Peginterferon alfa-2a alone, lamivudine alone, and the two in combination in patients with HBeAg-negative chronic hepatitis B. N Engl J Med. 2004 Sep 16;351(12):1206-17.

Locarnini S. Molecular virology and the development of resistant mutants: implications for therapy. Semin Liver Dis. 2005;25 Suppl 1:9- 19.

Zoulim F. Mechanism of viral persistence and resistance to nucleoside and nucleotide analogs in chronic hepatitis B virus infection. Antiviral Res. 2004 Oct; 64(1): 1-15.

Hussain M, et al. Lok AS. Mutations in the hepatitis B virus polymerase gene associated with antiviral treatment for hepatitis B. J Viral Hepat. 1999 May;6(3): 183-94. Review

Wright TL. Clinical trial results and treatment resistance with lamivudine in hepatitis B. Semin Liver Dis. 2004;24 Suppl 1:31-6. Review.

Dienstag JL, et al. Goldin RD, Heathcote EJ, Hann HW, Woessner M, Stephenson SL, Gardner S, Gray DF, Schiff ER. Histological outcome during long-term lamivudine therapy. Gastroenterology. 2003 Jan;124(l):105-17. Tang XR, et a^'l. Zhang JS, Zhao H, Gong YH, Wang YZ, Zhao JL. Detection of hepatitis B virus genotypes using oligonucleotide chip among hepatitis B virus carriers in Eastern China. World J Gastroenterol. 2007 Apr 7;13(13):1975-9.

Vernet G, et al. Tran N. The DNA-Chip technology as a new molecular tool for the detection of HBV mutants. J Clin Virol. 2005 Dec;34 Suppl l:S49-53.

Sertoz RY, et al. Erensoy S, Pas S, Akarca US, Ozgenc F, Yamazhan T, Ozacar T, Niesters HG. Comparison of sequence analysis and INNO- LiPA HBV DR line probe assay in patients with chronic hepatitis B. J Chemother. 2005 Oct;17(5):514-20.

Wang C, et al. Mitsuya Y, Gharizadeh B, Ronaghi M, Shafer RW. Characterization of mutation spectra with ultra-deep pyrosequencing: application to HIV-I drug resistance. Genome Res. 2007 Aug;17(8):1195-201. Epub 2007 Jun 28

TF Smith et al. and MS Waterman. Identification of common molecular subsequences. J. MoI. Biol. 147: 195-197, 1981.

O Gotoh. An improved algorithm for matching biological sequences. J. MoI. Biol. 162: 705-708, 1982.

SF Altschul, et al. W Gish, W Miller, EW Myers, DJ Lipman. Basic local alignment search tool. JMoI Biol. 215: 403-10, 1990.

N Eriksson, et al. L Pachter, Y Mitsuya, SY Rhee, C Wang, B Gharizadeh, M Ronaghi, RW Shafer and N Beerenwinkel. Viral population estimation using pyrosequencing. PLoS Comput Biol. 4(5): el000074. 2008.

A Agresti et al. and BA Coull. Approximate is better than "exact" for interval estimation of binomial proportions. The American Statistician 52(2): 119-126. 1998.

Keeffe EB, et al. Dieterich DT, Pawlotsky JM, Benhamou Y. Chronic hepatitis B: preventing, detecting, and managing viral resistance. Clin Gastroenterol Hepatol. 2008 Mar;6(3):268-74. Review.

Lok AS, et al. Zoulim F, Locarnini S, Mangia A, Niro G, Decraemer H, Maertens G, Hulstaert F, De Vreese K, Sablon E. Monitoring drug resistance in chronic hepatitis B virus (HBV)-infected patients during lamivudine therapy: evaluation of performance of INNO-LiPA HBV DR assay. J Clin Microbiol. 2002 Oct;40(10):3729-34.

Zoulim F. New nucleic acid diagnostic tests in viral hepatitis. Semin Liver Dis. 2006 Nov;26(4):309-17.

Hoffmann C, et al. Minkah N, Leipzig J, Wang G, Arens MQ, Tebas P, Bushman FD. DNA bar coding and pyrosequencing to identify rare HIV drug resistance mutations.iVucZeic Acids Res. 2007;35(13):e91. Epub 2007 Jun 18.

M. Margulies et al., Nature 437, 376 (2005).

Ronaghi M, et al. Uhlen M, Nyren P. A sequencing method based on real-time pyrophosphate. Science. 1998 JuI 17;281(5375):363, 365.

Ronaghi M, et al. Karamohamed S, Pettersson B, Uhlen M, Nyren P. Real-time DNA sequencing using detection of pyrophosphate release. Anal Biochem. 1996 Nov l;242(l):84-9

Langley, P. (1996) "Elements of Machine Learning", Morgan Kaufmann Ed

Mitchell, T. (1997). "Machine Learning", McGraw Hill Ed

Witten, I. & Frank, E. (2005) "Data Mining: Practical Machine Learning Tools and Techniques", Morgan Kaufmann Ed

Fattovich, G. 2003. Natural history of hepatitis B. J. Hepatol. 39 (Suppl):S50-58. DOPPIONE

Lok AS, et al. McMahon BJ. Chronic hepatitis B. Hepatology. 2007 Feb;45(2):507-39. Erratum in: Hepatology. 2007 Jun;45(6):1347.

Stuyver L, et al. Van Geyt C, De Gendt S, Van Reybroeck G, Zoulim F, Leroux-Roels G, Rossau R. Line probe assay for monitoring drug resistance in hepatitis B virus-infected patients during antiviral therapy. J Clin Microbiol. 2000 Feb;38(2):702-7.

Kim HS, et al. Han KH, Ahn SH, Kim EO, Chang HY, Moon MS, Chung HJ, Yoo W, Kim SO, Hong SP. Evaluation of methods for monitoring drug resistance in chronic hepatitis B patients during lamivudine therapy based on mass spectrometry and reverse hybridization. Antiυir Ther. 2005;10(3):441-9.

Hussain, M., et al. S. Fung, E. Libbrecht, E. Sablon, C. Cursaro, P. Andreone, and A. S. Lok. 2006. Sensitive line probe assay that simultaneously detects mutations conveying resistance to lamivudine and adefovir. J. Clin. Microbiol. 44:1094-1097.

Osiowy C, et al. Villeneuve JP, Heathcote EJ, Giles E, Borlang J. Detection of rtN236T and rtA181V/T mutations associated with resistance to adefovir dipivoxil in samples from patients with chronic hepatitis B virus infection by the INNO-LiPA HBV DR line probe assay (version 2). J CHn Microbiol. 2006 Jun;44(6):1994-7.

Libbrecht E, et al. Doutreloigne J, Van De Velde H, Yuen MF, Lai CL, Shapiro F, Sablon E. Evolution of primary and compensatory lamivudine resistance mutations in chronic hepatitis B virus-infected patients during long-term lamivudine treatment, assessed by a line probe assay. J Clin Microbiol. 2007 Dec;45(12):3935-41. Bartholomeusz A, et al. Tehan BG, Chalmers DK. Comparisons of the HBV and HIV polymerase, and antiviral resistance mutations. Antiυir Ther. 2004 Apr;9(2): 149-60. Review.

Das K, et al. Xiong X, Yang H, Westland CE, Gibbs CS, Sarafianos SG, Arnold E. Molecular modeling and biochemical characterization reveal the mechanism of hepatitis B virus polymerase resistance to lamivudine (3TC) and emtricitabine (FTC). J Virol. 2001 May;75(10):4771-9.

Allen MI, et al. Deslauriers M, Andrews CW, Tipples GA, Walters KA, Tyrrell DL, Brown N, Condreay LD. Identification and characterization of mutations in hepatitis B virus resistant to lamivudine. Lamivudine Clinical Investigation Group. Hepatology. 1998 Jun;27(6):1670-7.

Bozdayi AM, et al. Uzunalimoglu O, Tύrkyilmaz AR, Asian N, Sezgin O, Sahin T, Bozdayi G, Cinar K, Pai SB, Pai R, Bozkaya H, Karayalcin S, Yurdaydin C, Schinazi RF. YSDD: a novel mutation in HBV DNA polymerase confers clinical resistance to lamivudine. J Viral Hepat. 2003 Jul;10(4):256-65.

Sanger F, et al. Coulson AR. A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J MoI Biol. 1975 May 25;94(3):441-8.

Margulies, et al. M., M. Egholm, W.E. Altman, S. Attiya, J.S. Bader, L.A. Bemben, J. Berka, M.S. Braverman, Y.J. Chen, Z. Chen, S.B. Dewell, L. Du, J.M. Fierro, X.V. Gomes, B.C. Godwin, W. He, S. Helgesen, CH. Ho, G.P. Irzyk, S.C. Jando, M.L. Alenquer, T.P. Jarvie, K.B. Jirage, J. B. Kim, J. R. Knight, J. R. Lanza, J. H. Leamon, S.M. Lefkowitz, M. Lei, J. Li, K.L. Lohman, H. Lu, V.B. Makhijani, K.E. McDade, M.P. McKenna, E.W. Myers, E. Nickerson, J.R. Nobile, R. Plant, B.P. Puc, M.T. Ronan, G.T. Roth, G.J. Sarkis, J.F. Simons, J.W. Simpson, M. Srinivasan, K.R. Tartaro, A. Tomasz, K.A. Vogt, G.A. Volkmer, S.H. Wang, Y. Wang, M.P. Weiner, P. Yu, R.F. Begley, and J. M. Rothberg. Genome sequencing in microfabricated high-density picolitre reactors. Nature. 2005;437:376-80.

Nyren P. The history of pyrosequencing. Methods MoI Biol. 2007;373:l- 14.

Zoulim F. New nucleic acid diagnostic tests in viral hepatitis. Semin Liver Dis. 2006 Nov;26(4):309-17. Review.

TF Smith and MS Waterman. Identification of common molecular subsequences. J MoI. Biol 147:195-197, 1981

Eriksson, N., L. Pachter, Y. Mitsuya, S. Y. Rhee, C. Wang, B. Gharizadeh, M. Ronaghi, R. W. Shafer, and N. Beerenwinkel. 2008. Viral population estimation using pyrosequencing. PLoS. Comput. Biol. 4: e 1000074.

Wang C, Mitsuya Y, Gharizadeh B, Ronaghi M, Shafer RW. Characterization of mutation spectra with ultra-deep pyrosequencing: application to HIV-I drug resistance. Genome Res. 2007 Aug;17(8):1195-201.)

Huang X, et al. Madan A. CAP3: A DNA sequence assembly program. Genome Res. 1999 Sep;9(9):868-77.

Claims

1. Automated method for in vitro testing of mutations of HBV resistant to antiviral drugs to be carried out in biological samples comprising the following steps:

(i) carrying out the genome (or a portion thereof) sequencing of the HBV present in the biological samples from the patients; downloading the mutation data of the HBV genes which are in public or proprietary databases, where a reference population is needed; memorizing the data thus obtained;

(ii) performing the input of the information and of the data generated by the sequencing, and transformation thereof into standard electronic formats for the analysis; reading and memorization thereof by a parsing process; nucleotide alignment between the sequences obtained and with the sequences of the databases present in the memory, of the sequences obtained from patients; error correction; phylogenetic analysis; variation analysis of the sequences generated in step (i) to obtain prevalence and entropy data, with confidence intervals, regarding the observed variations;

(iii) translating the nucleotide sequences obtained in step (ii) into the corresponding aminoacid sequences; further correction of the errors occurred in said aminoacid sequences; analyzing the variation of the entropy and prevalence data with reference to the sensitivity to a drug or for a phylogenetic analysis by calculation of the confidence intervals of the mutation frequencies, also with the inclusion of mutations having a frequency above 1%;

(iv) extraction of the mutations and/or of the mutation profiles generated, which are associated with drug resistance or are due to a selection pushed by drug therapy, provided that statistic univariate or multivariate tests are carried out.

2. The method according to claim 1 further comprising a step (v) of correlating the mutations and/or the mutational profiles observed to acquired data on resistance to the treatment with antiviral drugs.

3. The method according to claim 2, wherein the drug resistance data are acquired by means of a phenotyping technique on at least one strain of the mutant viral HBV isolate.

4. The method according to claims 1-3, wherein the sequencing is carried out by ultra-deep pyrosequencing.

5. Mutations and/or mutational profiles of the HBV genome, wherein the mutations are indicated with the x_n_y code and are selected in the group: s:E_2_A; s:E_2_D; rt:D_2_G; rt:W_3_R; s:N_3_S; rt:G_4_E; rt:G_4_R; s:I_4_T; s:T_5_A; s:T_5_I; s:T_5_S; rt:A_7_D; s:G_7_R; rt:A_7_T; rt:A_7_V; rt:A_7_Y; rt:E_8_D; s:F_8_H; rt:E_8_K; s:F_8_P; s:F_8_R; s:L_9_R; rt:H_9_Y; s:G_10_A; s:G_10_K; rt:G_10_R; s:G_10_R; rt:E_ll_D; rt:E_ll_K; s:P_ll_L; rt:E_ll_Q; s:L_12_V; rt:H_13_L; rt:H_13_N; s:L_13_P; rt:H_13_R; rt:H_13_Y; s:V_14_A; s:V_14_G; rt:R_15_K; s:G_18_del; rt:R_18_K; rt:R_18_S; s:F_19_C; s:F_20_L; rt:P_20_S; s:F_20_S; s:L_21_S; rt:R_22_C; rt:R_22_G; rt:V_23_A; rt:T_24_A; s:R_24_K; s:R_24_L; s:R_24_S; s:I_25_T; rt:G_26_R; s:T_27_del; rt:V_27_F; s:T_27_I; s:I_28_del; s:I_28_L; s:P_29_del; s:P_29_L; rt:V_30_G; s:Q_30_H; s:Q_30_K; s:Q_30_W; rt:D_31_N; s:S_31_N; s:S_31_R; s:L_32_Q; s:D_33_A; s:D_33_L; s:S_34_L; s:W_35_F; s:W_35_G; rt:H_35_Y; rt:N_36_H; s:W_36_L; s:W_36_U; s:T_37_P; rt:T_37_S; s:T_37_S; rt:A_38_E; rt:A_38_P; rt:A_38_S; rt:A_38_T; s:L_39_D; s:L_39_del; s:L_39_P; s:N_40_del; s:N_40_F; s:N_40_H; s:N_40_S; s:N_40_T; s:F_41_L; s:L_42_G; s:L_42_R; s:G_43_del; s:G_43_M; s:G_43_R; s:G_44_E; s:E_44_R; s:S_45_A; s:T_45_A; s:S_45_ins; s:T_45_N; s:T_45_S; s:S_45_V; s:P_46_G; . s:T_46_P; s:P_46_V; s:V_47_A; s:V_47_del; s:V_47_E; s:V_47_G; s:V_47_I; s:V_47_ins; s:V_47_K; s:V_47_N; s:V_47_T; s:C_48_F; s:C_48_G; rt:Q_48_H; s:L_49_P; s:L_49_R; s:G_50_A; s:G_50_S; rt:S_50_W; rt:R_51_K; s:Q_51_L; s:Q_51_R; s:Q_51_V; s:N_52_D; rt:G_52_E; rt:G_52_R; rt:N_53_D; rt:N_53_I; rt:N_53_K; s:S_53_L; rt:H_53_S; rt:N_53_S; rt:N_53_V; rt:H_54_D; s:Q_54_H; s:Q_54_N; s:Q_54_R; rt:H_54_S; rt:H_54_T; rt:R_55_del; s:S_55_del; s:S_55_F; rt:R_55_H; rt:R_55_Q; rt:V_56_M; s:P_56_Q; s:P_56_S; s:P_56_T; s:T_57_A; s:T_57_I; s:T_57_S; s:S_58_C; s:S_58_F; rt:W_58_U; s:N_59_S; s:N_59_Y; s:H_60_Q; s:H_60_R; s:S_61_L; s:S_61_P; s:P_62_L; s:T_63_I; s:T_63_N; s:T_63_S; s:S_64_C; s:S_64_P; s:C_65_Y; s:P_67_Q; s:T_68_I; s:P_70_A; s:P_70_L; s:W_74_U; s:M_75_L; s:C_76_F; s:C_76_S; s:C_76_Y; rt:S_78_C; s:R_78_Q; s:R_79_C; s:R_79_G; s:R_79_H; s:R_79_L; s:R_79_S; s:F_80_C; rt:L_80_I; s:F_80_L; s:F_80_S; rt:L_80_V; s:I_81_H; s:I_82_L; rt:L_82_M; s:I_82_M; s:F_83_C; s:F_83_L; s:F_83_S; s:L_84_P; s:F_85_C; s:F_85_I; s:I_86_R; rt:A_87_E; rt:A_87_G; rt:A_87_T; rt:A_87_V; rt:F_88_S; rt:Y_89_S; rt:H_90_P; s:C_90_S; rt:I_91_F; rt:I_91_P; rt:I_91_V; s:I_92_L; s:I_92_M; rt:L_93_N; s:V_96_G; s:L_97_del; rt:A_97_S; rt:A_97_T; s:L_98_V; s:S_100_C; s:Y_100_C; s:Q_101_R; rt:V_103_I; s:M_103_I; s:M_103_L; rt:I_103_V; rt:G_104_S; s:P_105_A; s:P_105_S; s:A_105_V; rt:S_106_A; s:V_106_G; s:V_106_I; s:C_107_D; rt:G_107_R; s:C_107_R; s:C_107_Y; s:L_109_G; rt:S_109_P; rt:R_110_G; s:I_110_L; s:I_110_V; s:P_lll_L; s:P_lll_R; rt:V_112_A; s:G_112_A; s:G_112_E; rt:V_112_L; s:G_112_R; s:S_113_A; s:S_113_del; rt:A_113_G; s:S_113_T; rt:R_114_H; s:S_114_P; s:S_114_T; s:T_115_P; rt:L_115_V; s:T_116_S; s:S_117_R; s:S_117_T; s:T_118_A; rt:N_118_D; rt:N_118_S; rt:N_118_T; s:T_118_V; rt:N_118_Y; rt:R_120_K; rt:R_120_S; s:P_120_S; s:P_120_T; rt:I_121_N; rt:I_121_S; rt:I_121_V; rt:F_122_H; rt:F_122_I; s:R_122_K; rt:F_122_N; s:R_122_Q; rt:F_122_S; rt:F_122_V; rt:F_122_Y; rt:N_123_D; rt:N_123_T; s:T_123_V; rt:Q_125_H; s:T_125_M; s:T_126_I; s:T_126_S; rt:H_126_Y; s:P_127_L; rt:G_127_R; s:P_127_T; rt:T_128_A; s:A_128_E; rt:T_128_I; rt:T_128_N; s:A_128_T; s:A_128_V; s:Q_129_H; rt:M_129_L; s:Q_129_R; s:Q_129_U; s:G_130_del; s:G_130_E; s:G_130_R; s:N_131_A; s:T_131_A; rt:D_131_H; s:T_131_N; rt:D_131_S; s:K_133_I; s:M_133_I; s:K_133_L; s:M_133_R; s:K_133_T; s:M_133_T; s:F_134_C; s:Y_134_C; s:F_134_del; rt:D_134_E; s:Y_134_F; rt:D_134_G; rt:D_134_N; s:Y_134_N; rt:D_134_V; s:P_135_A; s:P_135_H; rt:H_135_T; s:S_136_L; rt:C_136_Y; s:C_137_G; rt:S_137_T; s:C_138_G; rt:R_138_K; rt:D_139_H; rt:D_139_K; rt:D_139_Q; rt:D_139_S; s:C_139_Y; s:T_140_del; s:I_140_S; s:T_140_S; s:K_141_del; rt:Y_141_F; rt:Y_141_Q; s:K_141_T; s:K_141_U; rt:V_142_E; rt:V_142_I; s:S_143_L; s:T_143_L; s:S_143_T; s:G_145_E; rt:L_145_M; s:G_145_R; s:N_146_H; s:N_146_T;s: C_147_Y; s:T_148_del; rt:Y_148_F; rt:K_149_H; s:C_149_R; s:C_149_Y; s:I_150_T; rt:F_151_L; s:P_151_L; rt:F_151_V; rt:F_151_Y; s:I_152_T; s:I_152_V; rt:R_153_Q; rt:W_153_R; rt:R_153_W; s:S_154_A; rt:K_154_E; rt:K_154_N; s:W_156_R; rt:L_157_P; s:F_158_L; s:G_159_A; s:G_159_E; rt:S_159_T; rt:H_160_R; s:K_160_R; s:Y_161_L; s:F_161_Y; s:L_162_I; s:W_163_C; s:W_163_U; rt:I_163_V; s:E_164_A; s:E_164_D; s:E_164_G; rt:L_164_M; s:E_164_V; s:L_165_U; s:A_166_V; s:V_168_D; rt:K_168_E; s:A_168_V; s:R_169_G; s:F_170_ins; s:F_170_S; s:F_170_T; s:W_172_C; rt:G_172_U; s:W_172_U; s:L_173_F; rt:V_173_L; s:L_173_P; s:S_174_N; s:S_174_T; s:L_175_F; s:L_175_S; s:L_176_R; s:V_177_A; s:V_177_E; s:V_177_M; s:P_178_Q; s:P_178_T; s:F_179_L; s:F_179_V; s:V_180_A; s:V_180_L; rt:L_180_M; s:Q_181_E; rt:A_181_S; rt:A_181_T; rt:A_181_V; s:W_182_U; s:F_183_C; s:F_183_L; s:V_184_A; rt:T_184_S; rt:A_186_T; s:S_187_del; rt:I_187_L; rt:I_187_T; rt:I_187_V; rt:C_188_R; s:T_189_I; s:V_190_A; s:V_190_G; s:W_191_U; s:L_192_F; s:L_192_I; rt:R_192_S; s:L_192_V; s:S_193_K; s:S_193_L; s:S_193_N; s:V_194_A; s:A_194_D; s:A_194_G; s:I_195_L; s:I_195_M; s:I_195_T; s:W_196_C; s:W_196_L; s:W_196_S; s:W_196_U; s:M_197_I; s:M_197_L; s:M_197_T; s:M_198_I; s:M_198_K; s:M_198_R; s:M_198_T; s:W_199_R; s:W_199_U; rt:L_199_V; s:Y_200_F; rt:A_200_G; s:Y_200_L; s:Y_200_N; s:Y_200_S; rt:A_200_T; rt:A_200_V; s:W_201_del; rt:F_201_U; s:W_201_U; s:G_202_A; rt:S_202_R; s:G_202_R; s:P_203_ins; s:P_203_L; s:P_203_Q; s:P_203_R; rt:Y_203_S; s:P_203_T; s:P_203_U; s:R_204_G; s:S_204_G; rt:M_204_I; s:R_204_K; rt:M_204_L; s:R_204_N; s:S_204_N; s:S_204_R; rt:M_204_V; s:L_205_G; rt:D_205_N; rt:D_205_Y; s:Y_206_C; rt:D_206_E; s:Y_206_F; s:Y_206_H; s:Y_206_P; s:Y_206_Q; s:Y_206_R; rt:D_206_U; s:R_207_C; rt:V_207_E; s:R_207_G; s:N_207_H; rt:V_207_L; rt:V_207_M; s:R_207_N; s:N_207_R; s:N_207_T; rt:V_208_A; rt:V_208_E; s:I_208_N; s:I_208_S; s:I_208_T; s:L_209_I; s:L_209_M; s:L_209_T; rt:L_209_V; s:L_209_V; rt:G_210_E; s:R_210_K; s:S_210_M; s:R_210_N; s:S_210_N; s:S_210_R; s:R_210_T; rt:A_211_D; rt:A_211_I; s:P_211_L; s:P_211_Q; s:F_212_del; s:F_212_L; rt:K_212_N; rt:K_212_R; s:I_213_F; s:L_213_I; s:I_213_M; s:L_213_M; rt:S_213_R; s:I_213_S; s:L_213_T; rt:V_214_A; s:P_214_H; s:P_214_L; rt:P_215_H; rt:P_215_T; s:L_216_C; s:L_216_F; rt:H_216_Q; s:L_216_U; rt:H_216_Y; s:L_217_E; rt:L_217_H; s:L_217_ins; rt:L_217_R; rt:E_218_D; rt:E_218_H; s:I_218_ins; s:I_218_S; s:I_218_T; s:F_219_I; s:F_219_S; s:F_220_C; s:F_220_del; rt:L_220_F; s:F_220_L; s:F_220_S; rt:L_220_T; s:C_221_G; s:C_221_R; rt:F_221_Y; s:C_221_Y; rt:T_222_A; s:L_222_del; s:L_222_P; rt:T_222_S; s:L_222_V; rt:A_223_P; s:W_223_R; rt:A_223_S; s:W_223_U; s:V_224_A; s:V_224_E; s:V_224_G; rt:V_224_I; rt:V_224_L; s:Y_225_C; s:Y_225_F; s:Y_225_H; rt:T_225_P; rt:T_225_S; s:Y_225_S; s:Y_225_U; rt:N_226_H; rt:F_227_H; rt:L_228_P; rt:L_229_M; rt:L_229_S; rt:L_229_V; rt:L_229_W; rt:S_230_C; rt:S_230_P; rt:S_230_Y; rt:L_231_H; rt:L_231_S; rt:G_232_V; rt:I_233_L; rt:I_233_M; rt:I_233_P; rt:I_233_T; rt:I_233_V; rt:H_234_L; rt:L_235_F; rt:L_235_U; rt:L_235_V; rt:N_236_K; rt:N_236_T; rt:N_236_V; rt:P_237_L; rt:P_237_Q; rt:P_237_T; rt:A_238_H; rt:A_238_K; rt:A_238_S; rt:K_241_R; rt:R_242_G; rt:G_244_R; rt:Y_245_H; rt:S_246_N; rt:S_246_P; rt:S_246_T; rt:N_248_H; rt:H_248_Y; rt:F_249_L; rt:M_250_I; rt:M_250_T; rt:Y_252_H; rt:V_253_I; rt:I_253_T; rt:G_255_R; rt:S_256_C; rt:C_256_G; rt:Y_257_G; rt:Y_257_H; rt:W_257_L; rt:Y_257_W; rt:S_259_L; rt:T_259_P; rt:T_259_S; rt:S_259_T; rt:Q_262_U; rt:H_264_Y; rt:I_266_T; rt:I_266_V; rt:Q_267_H; rt:Q_267_L; rt:Q_267_R; rt:Q_267_Y; rt:I_269_L; rt:I_269_T; rt:E_271_A; rt:E_271_D; rt:H_271_D; rt:H_271_E; rt:E_271_H; rt:E_271_L; rt:E_271_Q; rt:R_274_G; rt:R_274_Q; rt:R_274_S; rt:I_278_A; rt:V_278_I; rt:R_280_M; rt:D_283_E; rt:W_284_R; rt:K_285_Q; rt:V_286_L; rt:C_287_F; rt:Q_288_U; rt:R_289_S; rt:I_290_L; rt:I_290_N; rt:V_291_G; rt:V_291_M; rt:V_291_W; rt:G_292_A; rt:G_292_W; rt:L_293_V; rt:L_294_F; rt:L_294_U; rt:G_295_D; rt:F_296_L; rt:F_296_M; rt:F_296_S; rt:A_297_I; rt:A_297_T; rt:A_298_D; rt:A_298_V; rt:F_300_Y; rt:T_301_K; rt:Q_302_E; rt:C_303_F; rt:Y_305_A; rt:P_306_S; rt:P_306_T; rt:A_307_V; rt:L_308_E; rt:K_309_L; rt:P_310_L; rt:L_311_S; rt:A_313_T; rt:Q_316_H; rt:A_320_P; rt:F_321_T; rt:T_322_P; rt:T_322_S; rt:T_322_V; rt:F_323_L; rt:S_324_R; rt:P_325_H; rt:A_329_D; rt:A_329_T; rt:F_330_C; rt:L_331_V; rt:C_332_F; rt:C_332_N; rt:C_332_R; rt:C_332_S; rt:C_332_Y; rt:K_333_N; rt:K_333_T; rt:Y_335_F; rt:L_336_M; rt:N_337_D; rt:N_337_H; rt:N_337_T; wherein rt: reverse transcriptase gene (polimerase); s: gene codifying for the surface antigen; x = aminoacid in wild type (D or A genotype); n = positive integer indicating the position of the the aminoacid codon

(standard numbering); y = aminoacid substituting the wild type one (mutation); further, when y = U the mutation inserts a stop codon.

6. Mutations and/or mutation profiles according to claim 5, hwerein the mutationsa are selected in the group: rt:L_115_V; rt:F_122_S; rt:N_123_T; rt:T_128_I; rt:H_135_T; rt:V_142_E; rt:F_151_L; rt:S_159_T; rt:L_180_M; rt:T_184_S; rt:I_187_L; rt:I_187_T; rt:A_200_G; rt:M_204_I; rt:M_204_V; rt:K_212_N; rt:V_214_A; rt:L_229_M; rt:L_229_W; rt:Y_245_H; rt:S_246_T; rt:Y_257_H; rt:I_266_T; rt:E_271_Q; rt:I_290_N; rt:V_291_G; rt:L_294_F.

7. Mutations and/or mutational profiles of HBV genome detected by means of the automated method according to claims 1-4.

8. Use of the mutations and mutational profiles according to claims 5-7 for building a database of genomic mutations of HBV to be employed in the medical field, in particular for the diagnosis and treatment of viral forms resistant to drugs against the HBV.

9. Automated device for obtaining HBV genome mutations and/or mutational profiles resistant to antiviral drugs comprising: a unit for sequencing HBV genome and for downloading accessibile data from databases; a unit for processing, parsing, alignment, error correction, clustering, phylogenetic analysis and analysis of vatiation of data; a unit for extraction of mutations of the reverse transcriptase and/or the surface antigen from the sequences thus obtained; a unit for the local optimized alignment and error correction; a unit for evaluating the nucleotide and aminoacid variability, with calculation of confidence intervals of mutation prevalences; a unit for including mutations with frequence higher than 1%; a unit for extracting the mutations and/or mutational profiles thus generated.

10. Device according to claim 9, wherein the sequencing unit is an ultra-deep pyrosequencing unit.

11. The device according to claims 9-10, further comprising a unit for correlating the mutations found to drug resistance.

12. Use of the pyrosequencing technique for genotypic characterization of the viral population in an HBV infected patient.

13. Method for evaluating the effectiveness of antiviral therapy in an HBV infected patient, said method being carried out on a biological sample, preferably blood, of the patient and comprising the steps of: determining by means of the method according to claims 1-4 whether the sample comprises at least one nucleic acid codifying for HBV reverse transcriptase and/or surface antigen containing at least one mutation linked to drug resistance; and using the presence of said mutated nucleic acid for evaluating the effectiveness of antiviral therapy.

14. Method for determining the sensitivity to one or more drug(s) of a viral population present in an HBV infected patient, said method being carried out on a biological sample, preferably blood, of the patient and comprising the steps of: determing by means of the method according to claims 1-4 whether the sample contains at least one nucleic acid codifying for HBV reverse transcriptase and/or surface antigen containing at least one mutation linked to drug resistance; and using the presence of said mutated nucleic acid in choosing an antiviral therapy.

15. Method to select a therapy for an HBV infected patient, said method being caried out on a biological sample, preferably blood, of the patient, comprising the steps of: determining by means of the method according to claims 1-4 whether the sample comprises at least one nucleic acid codifying for HBV reverse transcriptase and/or surface antigen containing at least one mutation linked to drug resistance; and using the presence of said mutated nucleic acid for selecting the antiviral therapy.

16. Method for identifying drugs effective on HBV strains resistant to antiviral drugs therapy, said method comprising the steps of: obtaining at least one strain of wild type or engineered HBV comprising in its genome at least one of the mutations identified by the method according claims 1-4; determining the phenotypic response of the virus to the test drug; and using the phenotypic response thus obtained to ascertain the effectiveness of the tested drug.

17. HBV oligo- or polynucleotide sequence comprising mutations and/or mutational profiles according to claims 5-7 to be used in the medical field, in particular for the diagnosis of resistance to antiviral drugs.

18. Plasmid comprising the oligo- or polynucleotide sequence according to claim 17.

19. The plasmid according to claim 18 to be used as a cloning and expression vector.

20. Engineered cell comprising the vector according claim 19, in particular HepG2 cells.

21. Engineered HBV containing one or more resistance mutations to antiviral drugs identified by means of the method according to claims 1- 4.

22. A primer chosen from the following sequences:

SEQ ID No 1 HBV 1 FW CCTGCTGGTGGCTCCAGTT

SEQ ID No 2 HBV 1 RW AGAGAAGTCCACCACGAG

SEQ ID No 3 HBV 2 FW CCTGCTCGTGTTACAGGCG

SEQ ID No 4 HBV 2 RW CCGCAGACACATCCAGCG

SEQ ID No 5 HBV 3 FW CCGTGTGTCTTGGCCAAA

SEQ ID No 6 HBV 3 RW GACAAACGGGCAACATAC

SEQ ID No 7 HBV 4 FW GCTGCTATGCCTCATCTTCT

SEQ ID No 8 HBV 4 RW GAYGATGGGATGGGAATAC

SEQ ID No 9 HBV 5 FW GCACGACTCCTGCTCAAGG

SEQ ID No 10 HBV 5 RW CCCKACGAACCACTGAACAA

SEQ ID No Il HBV 6 FW GTATTCCCATCCCATCRTC

SEQ ID No 12 HBV 6 RW CGGTAWAAAGGGACTCAMG

SEQ ID No 13 HBV 7 FW TTGTTCAGTGGTTCGTMGGG

SEQ ID No 14 HBV 7 RW GGGTTAAATGTATACCCAVAG

SEQ ID No 15 HBV 8 FW CKTGAGTCCCTTTWTACCG

SEQ ID No 16 HBV 8 RW CKTGAGTCCCTTTWTACCG