FR3051482A1 - METHOD FOR DETERMINING THE PRESENCE AND QUANTIFICATION OF AT LEAST ONE MICROORGANISM IN A BIOLOGICAL SAMPLE - Google Patents

Abstract

The present invention relates to a method for determining the presence and quantification of at least one microorganism in a human biological sample comprising total nucleic acids, comprising the following steps: - extraction of the total nucleic acids from said biological sample - high sequencing flow rate of said total nucleic acids - computer processing of sequencing data - filtering of non-human sequences by alignment of sequences with reference genome sequences of at least one microorganism of at least one reference sample - calculation of the sequencing depth of the reference genome of at least one microorganism - the determination of an indicator of the quantification of at least one microorganism that is a function of said depth of sequencing of the reference genome. The present invention also relates to equipment and kits for the implementation of said method, for prenatal diagnosis or cancer.

Description

Method for determining the presence and quantification of at least one microorganism in a biological sample

Field of the invention

The present invention relates to the field of contaminant detection in a biological sample by high throughput sequencing and assay of said contaminants in the biological sample.

More particularly, the invention relates to the detection and quantification of microorganisms, for example in a human biological sample, in particular plasma, whose concentration in microorganisms is not known beforehand.

Background of the invention

High throughput sequencing techniques (also called next-generation sequencing techniques) are of particular interest in genomic studies, particularly in physiopathology. Indeed, these techniques make it possible to sequence the nucleic acids and in particular the whole genome or the entire exome of a biological tissue or to more specifically sequence a panel of genes of interest. It is possible from sequenced genomic data to study the microbiome of the tissues studied.

In 2008, the NIH initiated the Human Microbiome Project (HMP), which aimed to characterize the microbiome of various human biological tissues (NIH HMP Working Group et al., 2009). Several studies then made it possible to highlight the impact of the micro-organism in various pathological situations. Thus it is possible to associate the presence of certain microorganisms in a tissue to the emergence of a pathology. For example, during pregnancy, many infections (parasitic, bacterial, viral ...) can have serious consequences for the mother and the fetus. Among the most common are infections with cytomegalovirus (CMV) or rubella virus (Pereira et al., 2005). Screening for these infections in pregnant women could lead to better management of at-risk pregnancies. Similarly, it has been shown in transplant rejection cases that identification of viral-derived sequences may be associated with increased risk of graft rejection (De Vlaminck et al., 2015). Finally, some viral infections can be associated with particular oncogenic processes. This is notably the case of the HPV-16 virus associated with the emergence of a cancer of the cervix or hepatitis B virus that can lead to the appearance of hepatic carcinomas (Blackadar, 2016). In all cases, the detection of specific infections, as well as the quantification of these infections, can improve the diagnosis and monitoring of patients with various pathologies.

State of the art

State-of-the-art techniques for detecting biomarkers of the health status of a fetus in a maternal biological sample are known.

The patent application WO2015 / 070086 describes a method for analyzing the microbiome in a subject comprising high-throughput sequencing of a sample of the patient, bioinformatic analysis of the sequencing data and determination of the presence of a microbial sequence.

Patent Application WO2014 / 019275 discloses a method of identifying biomarkers of fetal health status from a foreign organism such as a virus, bacteria, fungus and parasite, and its uses for determining non-invasive state of fetal health. This patent application also relates to a system and a kit that are used in the process.

Non-invasive biomarker detection methods are available in the state of the art based on quantitative PCR (Polymerase Chain Reaction) DNA detection techniques specifically targeting a DNA sequence originating from a particular microorganism. . There are also techniques for immunological detection of an infection, in particular by quantifying immunoglobulins.

There are also in the state of the art biomarker detection methods that require an invasive step that may represent risks for the patient. This is the case, for example, with amniocentesis, a technique widely used to collect amniotic fluid in order to establish a battery of tests to evaluate the state of health of the fetus in pregnant women. However, amniocentesis is not without risk for the pregnant woman and the fetus since it can lead to the loss of the baby. Similarly, biopsies, for example to evaluate the carcinogenic nature of a tumor, constitute a surgical act that can be dangerous for the patient. The tests performed on these samples can be of different natures and more or less sensitive.

Disadvantage of the prior art

The methods of the prior art and in particular of the application WO2014 / 019275 only make it possible to determine the presence or absence of a contaminant from the sequencing data of a biological sample. These methods give no information on the concentration of contaminant in the sample. It is also not possible to follow the evolution of this concentration over time.

The method of WO2015 / 070086 does not include an optimization of the detection of microorganisms within a subject. Indeed, a number of reads can non-specifically align with the reference genome at low genomic region level and be the source of false positives.

This document also does not establish beforehand a reference matrix on a plurality of reference samples whose concentration in microorganism is known, and to record a matrix corresponding to the depth determined by said characterization, and the concentration in micro corresponding known organism. This document establishes at most a relative quantification but no absolute quantification of the concentration of microorganism.

Other non-invasive quantitative PCR detection techniques, for example, generally have a high detection threshold and are generally long and non-exhaustive. Likewise, immunological techniques most often have a low specificity for detecting microorganisms. These techniques generally do not allow dating the infection in the sense that a positive result can result from a previous infection.

The immunological detection techniques of a microorganism can also be invasive for the patient, for example in the case of amniocentesis in pregnant women. These actions are not without consequences and they can represent risks for the patient and / or the fetus. Similarly, the detection threshold by these techniques is sometimes high which does not detect low levels of microorganisms.

In all cases, the concentration of microorganisms quantified by these techniques is not always very precise, which does not allow to precisely adjust the treatments.

There is therefore a real need to accurately detect and quantify several microorganisms in the same biological sample, whose quantity of said microorganisms is not known, in a safe, fast, precise, sensitive and with a very high quantification threshold. low.

Solution provided by the invention

The present invention proposes to overcome the disadvantages of the prior art by a method for determining the presence and quantification of at least one microorganism in a human biological sample comprising total nucleic acids, comprising the following steps: extraction total nucleic acids of said biological sample - high throughput sequencing of said total nucleic acids - computer processing of the sequencing data of counting the total number of sequences or filtering said sequences according to a quality score o filtering human sequences by alignment sequences with sequences of the human reference genome characterized in that said computer processing steps further comprise: - an initial calibration step of characterizing a plurality of reference samples whose concentration in microorganism is known, and to save a matrix corresponding to the depth determined by said characterization, and the corresponding known microorganism concentration - a step of masking from 0 to 5% of the nucleotides of the reference genome of at least one microorganism relative to the total nucleotides of said reference genome, said masked nucleotides corresponding to low complexity sequences - filtering the sequences of at least one microorganism by alignment of non-human sequences with the sequences of the reference genome of at least one microorganism from minus one reference sample - the calculation of the sequencing depth of the reference genome of at least one micro-organism - the validation of the homogeneity of the genome coverage of at least one micro-organism consisting of calculating the standard deviation of the sequencing depth of the genome to the reference of said genome of at least one microorganism - the determination of a quantifying at least one microorganism that is a function of said depth of sequencing of the reference genome.

The present invention provides a method for the rapid and accurate detection of microorganisms present in human biological samples, making it possible to demonstrate targeted deregulation of the microbiome within a biological tissue. The present invention also makes it possible to precisely quantify the microorganism in a biological sample. The present invention enables absolute detection and quantification of the microorganism concentration. The process according to the invention is very sensitive and makes it possible to detect at least one microorganism even if this or these latter are weakly present in the sample. The detection and quantification threshold of the process of the invention is very low. The detection threshold of the process according to the invention is below the detection threshold of commonly accepted techniques such as immunoassays. The quantification threshold of the process according to the invention is below the threshold of quantification of commonly accepted techniques such as quantitative PCR. The quality of the results obtained by the present method is increased compared to the solutions of the prior art. It is understood that the invention allows the detection of one or more microorganisms in the same sample of which the quantity of said microorganisms is not known beforehand. This quantification is absolute, in particular thanks to the initial calibration step of characterizing a plurality of reference samples whose concentration in microorganism is known, and to register a matrix corresponding to the depth determined by said characterization, and the corresponding known microorganism concentration.

The combination of the steps of the method according to the invention and in particular the steps of initial calibration, masking, filtering and validation of the homogeneity of coverage makes it possible to obtain reproducible, reliable results of very good quality - superior to the processes of the invention. state of the art.

Advantageously, the step of calculating the sequencing depth of the reference genome of at least one microorganism comprises a depth normalization treatment with respect to the total number of sequences resulting from the sequencing.

The method according to the invention optimizes and increases the amount of information resulting from the sequencing of human biological samples. The process according to the invention is rapid. The process according to the invention makes it possible to increase the quality of the results obtained. It also allows an absolute quantification of the concentration of microorganism.

Alternatively, the method according to the invention uses the sequencing data of human biological samples taken for other purposes, for example, for the screening of fetal aneuploidies or suspected cancerous biopsies. Thus the method according to the invention maximizes possible diagnoses from biological samples.

According to a second aspect, the invention relates to equipment for determining the presence and quantification of at least one microorganism in a biological sample comprising a computer controlled by a computer program for carrying out treatments of: initial calibration of characterizing a plurality of reference samples whose concentration in microorganism is known, and recording a matrix corresponding to the depth determined by said characterization, and the corresponding known microorganism concentration - masking from 0 to 5% of the nucleotides of the reference genome of at least one microorganism relative to the total nucleotides of the said reference genome, the said masked nucleotides corresponding to sequences of low complexity - filtering the sequences of at least one microorganism by alignment non-human sequences with sequences of at least one reference genome e of said microorganism of at least one reference sample - calculation of the sequencing depth of said reference genome - validation of the homogeneity of the genome coverage of at least one microorganism consisting in calculating the difference -type of the sequencing depth of the genome to the reference of said genome of at least one microorganism - determination of a microorganism concentration indicator according to said sequencing depth of the reference genome.

According to a third aspect, the invention relates to a computer program for the determination of the presence and quantification of at least one microorganism in a biological sample controlling the realization of the treatments of: - initial calibration of characterizing a plurality of reference samples whose concentration in microorganism is known, and to record a matrix corresponding to the depth determined by said characterization, and the corresponding known microorganism concentration - 0-5% masking of the nucleotides of the genome of reference of at least one microorganism relative to the total nucleotides of said reference genome, said masked nucleotides corresponding to sequences of low complexity - filtering the sequences of at least one microorganism by alignment of non-human sequences with the sequences of the reference genome of said microorganism of at least one sample of r reference - calculating the sequencing depth of the reference genome - validating the homogeneity of the genome coverage of at least one microorganism by calculating the standard deviation of the genome sequencing depth at reference said genome of at least one microorganism - determining a microorganism concentration indicator according to said sequencing depth of the reference genome.

According to a fourth aspect, the invention relates to a kit for determining the presence and quantification of at least one microorganism in a biological sample according to the above method comprising: - a plurality of tubes each containing a dilution of at least one DNA sequence of a reference genome of at least one microorganism in a control DNA - a control DNA containing tube - the nucleotide sequence of the reference genome of at least one microorganism of which 0 to 5% of the nucleotides are masked, said masked nucleotides corresponding to sequences of low complexity. In one embodiment, the DNA sequence of a reference genome of at least one microorganism is the sequence SEQ ID 1.

In a preferred embodiment, the invention relates to the application of the method for determining the presence and quantification of at least one microorganism in a biological sample to the determination of the concentration of a parasite.

In one embodiment, said parasite is a prokaryotic organism. Said prokaryote may be selected from bacteria or archae.

In another embodiment, said parasite is a virus.

The virus may be a DNA virus belonging to the family of herpesviridae, papillomaviridae, parvoviridae or any other family of DNA viruses.

The virus of the family Herpesviridae can be a cytomegalovirus, an Epstein-Barr virus, a varicellovirus, a simplex virus, a herpes virus type 8 or any other virus.

In another embodiment, said parasite is a eukaryotic organism.

In another embodiment, the invention relates to the application of the method for determining the presence and quantification of at least one microorganism in a biological sample for prenatal diagnosis, characterized in that said parasite is a cytomegalovirus.

In another embodiment, the invention relates to the application of the method for determining the presence and quantification of at least one microorganism in a biological sample for the diagnosis of a cancer induced by the Epstein Barr virus and / or monitoring an anticancer treatment characterized in that said parasite is an Epstein Barr virus.

In another embodiment, the invention relates to the application of the method for determining the presence and quantification of at least one microorganism in a biological sample for monitoring a graft, characterized in that said parasite is a cytomegalovirus virus.

According to another aspect, the invention relates to a kit for prenatal diagnosis using the method described above, characterized in that said sequence of the reference genome of at least one microorganism is a cytomegalovirus of which up to 0, 06% of the nucleotides are masked. Advantageously, said reference genome sequence of at least one microorganism is the sequence of the reference genome of a cytomegalovirus of which up to 0.06% of the nucleotides are masked, this sequence corresponding to the sequence SEQ ID 1.

In another aspect, the invention relates to a kit for monitoring a graft implementing the method described above characterized in that said reference genome sequence of at least one microorganism is a cytomegalovirus.

According to another aspect, the invention relates to a kit for the diagnosis of a cancer induced by the Epstein Barr virus and / or the monitoring of an anticancer treatment implementing the method previously described, characterized in that said genome sequences reference of at least one microorganism is an Epstein Barr virus.

Description

The present invention will be better understood in the light of the detailed description of the invention and nonlimiting examples of embodiment.

Figure 1 shows a diagram of the method according to the invention.

Figure 2 shows the distribution of bases according to their PHRED quality score for a sample.

Figure 3 shows the average distribution of nucleotide sequences from whole genome sequencing.

Figure 4 shows the number of viral sequences aligned on the reference genome of the masked CMV (CMV_NS) or conventional (NC_006273).

Figure 5 shows the comparison of the results obtained after filter or not human nucleotide sequences.

Figure 6 shows the performance comparison for alignment on the viral genome of the nucleotide sequences of 5 samples without (top) or with (bottom) a first step filtering the nucleotide sequences aligning with the human genome.

Figure 7 shows the objective number of sequences aligned on the CMV reference genome (CMV_NS) or normalized as a function of the number of total sequences of the sample in a sample infected with CMV.

FIG. 8 presents an estimation of the concentration by contaminating samples from the standardized mean depth and the correlation between the theoretical contaminant concentration of the sample and that estimated by the method according to the invention.

Figure 9 shows an exemplary embodiment of the invention for the detection of a viremia in plasma samples.

Figure 10 shows the profile of a positive sample according to the method of the invention and having a positive viremia for CMV.

Figure 11 shows the standardized mean depth of 4 samples (abscissa) for 9 microorganisms studied (family Herpesviridae). Definitions:

Human biological sample: Liquid or solid biopsies of human tissues, cells isolated from liquid or solid biopsies, biological fluids such as plasma or cerebrospinal fluid.

Fasta: Text file format for storing biological sequences such as nucleotide sequences of reference genomes.

Fastq: Text file format for storing sequences and their associated quality score such as nucleotide sequences from next-generation sequencing.

Microbiome: The microbiome includes all microorganisms that predominate or are permanently adapted to the surface and inside of a living organism. This term also refers to the sum of genomes of microorganisms living in or on an animal or plant organism. Pathological states of the animal or plant organism can be attributed to an imbalance of its microbiome. ORF (Open Reading Frame): Open reading phase is a region of the genome that can string for a protein. It is defined by the presence of an initiator codon and a stop codon which delimit a coding region surrounded in certain cases by regulatory sequences.

Sequencing Depth: The average number of times the genome is covered (often expressed in genome equivalents). In other words, the sequencing depth is defined as the number of sequences that cover a particular genomic region.

Read: Nucleotide sequence obtained after next-generation sequencing Sequencing: Consists of determining the order of nucleotides of a given nucleic acid sequence according to different methods. Next Generation Sequencing (or High Throughput Sequencing): Consists of massively and parallelizing the order of nucleotides of a large number of nucleic acid sequences within a sample. Typically this method allows the sequencing of the genome or transcriptome of a biological sample.

Quality score (Phred score): Quality score assigned to each nucleobase obtained after next generation sequencing. They make it possible to determine the accuracy of each nucleotide of a biological sequence stored in a fastq file.

Sample preparation

The nucleic acids of the biological samples are extracted according to conventional techniques well known to those skilled in the art. The nucleic acid extraction step is adapted to each biological sample and allows lysis of the cells to release the nucleic acids which are then purified and fragmented if necessary to allow their sequencing by next generation sequencing techniques. For example, the plasma is harvested after centrifugation of a blood sample and corresponds to the supernatant. This liquid phase contains no or few blood cells. The DNA circulating in the plasma is then extracted from this biological sample. In the case of circulating plasma DNA, no fragmentation step is necessary. Sequencing:

The samples are then sequenced by high-throughput sequencing techniques, for example sequencing in single-end or paired-end, long-reads or short-reads. This sequencing can be done on platforms well known to those skilled in the art, such as Illumina®, Roche® or IonTorrent® platforms. For example, in the case of sequencing according to Illumina® technology, adapter sequences are added at each end of the DNA fragments. These adapters are different for each sample and allow their identification after sequencing of several samples on the same chip. The addition of the adapters is followed by clonal amplification of the DNA fragments (by bridging or emulsion PCR, for example) and then by the sequencing step on the chosen automaton (for example HiSeql500 ™).

Pretreatment of the sequencing data: - Counting of the total number of sequences:

The sequencing data is stored in fastq files in the form of nucleotide sequences whose size depends essentially on the sequencing technology used, which is associated with a quality score called the Phred score. The number of sequences contained in each fastq file corresponds to the total number of sequences obtained after next generation sequencing. - Filtering of nucleotide sequences according to their quality:

The nucleotide sequences are then filtered on their quality. Only nucleotide sequences of sufficient quality are conserved (Figure 2). Similarly, the PCR duplicates are filtered during this step. - Filtering human sequences by alignment of sequences with sequences of the human reference genome:

A first alignment on the human reference genome GRCh38 is performed in order to filter the nucleotide sequences of human origin. Depending on the technology and the sequencing platform used, the alignment tool used should be adapted. Indeed, the algorithms used by the different aligners available are dependent, in particular on the size of the sequences, but also on their type of sequencing (in paired or single-end) and must be chosen according to these different parameters. For example, the Bowtie tool allows efficient and fast alignment of small (less than 50bp) sequences from whole genome sequencing (Langmead et al., 2009). In the case of sequencing data obtained after single-end sequencing of 26bp, sequences that align without variants and less than 5-fold on the human reference genome GRCh38 are considered nucleotide sequences of human origin. In order to accelerate this step, the best sequence alignment on the human reference genome may not be reported (Table 1).

Table 1: Example of Nucleotide Sequence Alignment Parameters on the GRCH38 Human Reference Genome with the Bowtie Alignment Tool for Illumina Sequencing Data Obtained on HiSeqlSOO ™ as a 26pb Single-End

In this example, on average, after whole genome sequencing, about 19% of the total sequences of sufficient quality do not align with the reference GRCH38 genome (FIG. 3).

Identification of the specific sequences of the contaminating DNAs - Genomes of references:

On certain particular genomic regions the non-specific alignment of many nucleotide sequences is observed. The number and size of these particular regions within a reference genome are independent of the size of the reference genome. This step is not mandatory for all reference genomes and depends mainly on the sequence of the reference genome. On the other hand, a maximum of 5 nucleotides per hundred nucleotides of the reference sequence can be modified at this stage in order to ensure sufficient alignment sensitivity for the rest of the analysis (modification of 0% to 5% of the nucleotides) .

For example, on a panel of 60 DNA viruses, 9 present a non-specific alignment of 4 to 173 sequences (Table 2).

Table 2: Viral genomes with low complexity nucleotide regions resulting in non-specific alignment of nucleotide sequences

Some particular genomic regions are generally composed of low complexity DNA sequences that correspond to low nucleotide diversity in general, one or two bases are overrepresented (eg AAAAAAAAAAATAAAAAAT). Regions of low complexity often cause non-specific alignments. Such regions may also include repetitions of patterns (sequences of some bases).

The nucleotide regions resulting in nonspecific alignments are thus masked, in order to allow a better homogeneity of the alignment on the reference genome and a decrease of the background noise observed for the negative samples. This method makes it possible to reduce the number of non-specific aligned sequences. For example, for the CMV virus (HSV-5), the step of masking the 4 low complexity regions passes through the modification of 155bp of the reference genome which become N (SEQ ID 1). In this case, the masking affects 0.06% of the nucleotides of the CMV genome.

This step makes it possible to reduce the number of sequences counted for the negative samples to a basal level close to zero and thus to discriminate more easily the low concentration samples in microorganisms and the negative samples (FIG. 4). - Alignment on the reference genome of microorganisms:

Nucleotide sequences that do not align with the human genome, whose origin is unknown, are then aligned with the target reference genome (s): parasite, viral, prokaryotic or eukaryotic microorganism genomes. Only sequences that align unambiguously are preserved. The number of variants tolerated by sequences during the alignment of this sequence is dependent on the microorganism studied and its gene variability and the sequencing platform used. For example, in the case of a non-variable micro-organism, such as DNA viruses sequenced on a platform Illumina HiSeql500 ™ single-end 26pb a maximum of 2 variants is tolerated for the following analysis (Table 3) .

Table 3: Example of alignment parameters of the nucleotide sequences on the microorganism reference genome with the Bowtie alignment tool for Illumina sequencing data obtained on HiSeqlSOO ™ 26pb single-end - Impact of the presence filter human nucleotide sequences on the detection of microorganisms (contaminating nucleic acid):

The first sequence alignment step on the human reference genome is performed on the latest available human reference genome: GRCh38 (Genome Reference

Consortium Human genome build 38) updated in 2013. The prior elimination of human sequences makes it possible to reduce the number of non-specific alignments. For example, the number of nucleotide sequences aligning with the CMV genome observed after the two alignment steps is less than that observed after direct alignment with the viral reference genome (FIG. 6).

Quantitative Analysis of Results - Validation of Total Sequencing Depth:

A first validation filter of the sequencing step is set up based on the minimum number of sequences that must be obtained to ensure a sufficient depth of sequencing to allow further analysis. In addition, a sufficient percentage of sequences aligning with the human reference genome is necessary in order to validate the smooth sequencing.

In the case of sequencing as previously described (Illumina HiSeql500 ™ single-end 26pb) a threshold of 10 million minimum sequences must be obtained (0.07X in human genome equivalent) to ensure robust and reproducible detection specific sequences of the microorganisms studied (Table 4). Thus, the method according to the invention requires a very low depth of sequencing of the human genome to sequence a maximum of samples simultaneously to minimize sequencing times and costs.

Table 4: Example of Sequencing Quality Filter Parameters for Illumina Sequencing Data Obtained on HiSeqlSOO ™ 26pb Single-End - Validation of Alignment Homogeneity:

An optional filter is used to identify biased alignments by checking if the genome is uniformly covered by the aligned interest sequences. This verification of the homogeneity of the alignment is performed by calculating the standard deviation of the sequencing depth of the genomic regions present uniformly in the sample.

Typically, for microorganisms having a DNA genome, this step can be performed on the entire genome of the microorganism. On the other hand, for microorganisms with an RNA genome, such as RNA viruses, it may be necessary to consider only those intergenic regions not subject to expression variations for this validation.

The depth of sequencing is defined as the number of sequences that cover a particular genomic region. The average depth of depthmean is defined as:

nb reads · * number of sequences aligned on the genome lengthreads: number of nucleotides of the genome size sequences: number of nucleotides of the reference genome The standard deviation of the sequencing depth of the genome is equal to:

depth ±: depth for a given genomic region on the reference genome depthmean: average depth over the entire genome reference sizegenome: number of nucleotides of the reference genome

As the depth follows a Poisson distribution, the theoretical standard deviation is equal to the root of the mean (depthmean). Finally, the parameter allowing to estimate the homogeneity of the alignment is a standardized (centered) and asymmetric standard deviation:

σ: standard deviation of the depth over the entire reference genome depthmean: mean depth over the entire reference genome genome size: number of nucleotides of the reference genome

Finally :

depth ±: depth for a given genomic region on the reference genome depthmean: average depth over the entire reference genome nbreads: number of sequences aligned on the genome lengthreads: number of nucleotides of the sequences Based on the results of good sequencing and poor quality, a threshold has been established for which it is possible to determine the quality of sequencing. Thus, a standard deviation below this threshold makes it possible, independently of the reference genome, to affirm that the alignment is good.

Quantitative analysis of the results: - Normalization of the number of sequences of microorganisms:

The number of specific sequences of each microorganism identified in a sample is dependent on the size of the reference genome of the organism studied. The more the microorganism studied has a large genome, the greater the probability of finding sequences of this microorganism in the library. In order to be able to quantify each microorganism independently of their size, we will not consider the number of sequences counted but the average depth of sequencing of the target genome (depthmean) as previously defined:

nbreads: number of sequences aligned on the genome lengthreads: number of nucleotides of the sequences taillegénome: number of nucleotides of the reference genome

The number of sequences counted is directly proportional to the total sequencing depth obtained at the scale of the total sample. The greater the number of sequences resulting from the sequencing, the more the number of sequences of microbial origin will be important for a sample having the same infectious load (FIG. 7).

The number of sequences is therefore normalized with respect to the number of total sequences:

nbreads: number of sequences aligned on the genome nbreadsTotai: number of total sequences resulting from the sequencing having passed the quality filter c: constant

The constant c finally makes it possible to relate the value of standardized depth to an interpretable scale:

readsnorm: number of sequences aligned on the genome after normalization lengthreads: number of nucleotides of the genome size sequences: number of nucleotides of the reference genome c: constant Analysis of cohorts of positive samples, that is to say infected with a micro- organism, and uninfected negative samples allows by a ROC analysis to determine a standardized average depth of sequencing of the target genome for which it is possible to determine whether the sample is infected or not. So if depthnorm is above the threshold then the sample will probably be infected by the microorganism of interest. The expected false-positive and false-negative rate is determined by the sensitivity and specificity results of the threshold determined by the ROC analysis. - Detection of an infection:

Screening for infection by a microorganism is made from the average depth of sequencing of the genome of the microorganism of interest, calculated and standardized as described above. The z-score is a statistical test to highlight a significant difference in a value within a population. A z-score greater than 3 makes it possible to identify samples for which, statistically, the number of nucleotide sequences aligning with the target genome is different from that of the population studied. The z-score is calculated from the mean (mean) and the absolute deviation of the standard deviation (StDev) from the number of aligned and normalized sequences of all the samples studied:

readsnorm: number of sequences aligned on the genome after normalization. - quantification of the infection: From increasing ranges of infected samples, sequenced beforehand, it is possible to perform a regression to estimate the infection rate of the samples tested. In general, these ranges, of known microbial sequence concentrations, must be sequenced at least in duplicate. A minimum of 4 range points is required for this step.

The normalized mean depth is calculated for each range point and is correlated linearly with the range point contaminant concentration (Figure 8). A linear regression is then performed to determine the equation for linking the infection rate (contaminant concentration) and the normalized average depth of genome sequencing (depthnorm):

depthnorm: standardized mean depth over the entire reference genome k: constant

Embodiment No. 1: Detection and Quantification of Cytomegalovirus (CMV) Plasma in Pregnant Women

CMV belongs to the family of herpesviridae. It is a double-stranded DNA enveloped virus of about 240kb. The infection is latent with maintenance of the viral genome as an episome in macrophages or endothelial cells (Bolovan-Fritts et al., 1999). Its transmission is interhuman and is mainly by contact between the mucous membranes. During the infection process, transmission of the virus in the host is carried out by intercellular contacts, in particular through circulating macrophages.

Primary CMV infection affects almost 50% of the world's population and is asymptomatic in most cases. CMV can be reactivated during the life of an individual. Similarly, infections from different viral strains can also be observed. Among the best known, 4 non-genetically engineered strains were isolated and fully sequenced (Dolan et al., 2004) (Table 5).

Table 5: Description of the main CMV reference strains (NCBI®)

Among the strains isolated after prolonged cell culture, deletions were observed that could explain their resistance and survival in culture. Finally, the reference strain considered to be the closest to the primary CMV observed in humans and the Merlin strain. CMV infection during pregnancy is mostly the result of a primary infection in an HIV-negative woman. More rarely, infection can result from secondary infection or viral reactivation in HIV-positive women. The clinical manifestations are in all cases non-existent or minor (flu syndrome).

Seronegative pregnant women in early pregnancy are therefore the main subjects at risk. In France, several epidemiological studies have shown that 50% to 55% of women of childbearing age have a negative CMV serology. In these women, the risk of infection during pregnancy is about 1%. The incidence of secondary infections or viral reactivations is poorly known (Adler, 2011).

In case of infection in the mother, the virus can infect the fetus via the fœto-placental barrier. The congenital infection of the fetus is defined by the detection of the virus in the circulation of the newborn in the first 3 weeks of life, and intervenes in 40% to 50% of the cases of primary infection during the pregnancy. In the end, the prevalence of CMV at birth is about 0.5% to 1%. The magnitude of complications resulting from congenital infection is still poorly defined. Symptoms can occur at birth in 10 to 20% of infected infants or in the first year of life in up to 30% of asymptomatic infants born at birth. Several types of neurological and / or sensory sequelae are often found in symptomatic children (Table 6). In 5% of cases, CMV infection at birth results in the death of the newborn (Benoist et al., 2013, James and Kimberlin, 2016).

Table 6: Description of the main sequelae observed in neonates with congenital CMV infection (James and Kimberlin, 2016)

The primary infection in the mother is all the more frequent as the age of gestation increases. In addition, it has been shown that the risk to the fetus increases when primary infection occurs in early pregnancy (before the first half, 26% of fetuses will present a severe pathology compared to only 6.2% during the second and third semesters of pregnancy) (Daiminger et al., 2005, Liesnard et al., 2000). To date, in France, only four infectious diseases are the subject of mandatory prenatal screening programs: toxoplasmosis, rubella, syphilis and hepatitis B. Although the vertical transmission of CMV is the main infectious cause of malformation congenital, prenatal CMV screening is not performed systematically. However, despite the absence of a recommendation, prenatal serological screening for CMV during pregnancy is being carried out more and more frequently.

During pregnancy, the detection of infection in the mother can be achieved through serological tests: assay of immunoglobulin type M and measurement of the avidity of immunoglobulin type G by Elisa tests. The immunoglobulin assay does not accurately identify primary infections because they may persist long after infection. Moreover their interpretation remains difficult, in particular because of a lack of standardization of the proposed tests and the serological status of the infected subjects over time. According to the High Health Authority, the date of infection is only possible in 75% to 80% of cases. In addition, this test does not identify secondary infections.

Ultrasound monitoring can identify non-specific embryonic developmental abnormalities of congenital CMV infection. However, in nearly half of the cases with sequelae, no signs are observed during pregnancy, hence the importance of detecting CMV.

Currently, confirmation of fetal CMV infection can be achieved through an invasive procedure. The presence of particles or viral DNA in the amniotic fluid is demonstrated by cell culture or PCR techniques and can then diagnose an infection in the fetus. However, invasive procedures are dangerous and their specificity, as well as their sensitivity, remain limited. False-negative results may be due to amniocentesis performed too early during fetal development (fetal urination function appears only after 20 weeks of pregnancy) and too soon after infection (6 to 8 weeks for CMV is excreted in the urine). In addition, false positives can be the result of a contamination by the maternal blood intervening following the invasive gesture (amniocentesis).

The symptomatology of the infection in the adult is not specific (flu syndrome) and does not allow to make the diagnosis of the CMV infection. Current serological tests do not identify all active infections in pregnant women, particularly in the case of secondary infections or viral reactivation, and are not accurate. In addition, ultrasound alone can not be used to estimate the risk of congenital fetal infection. In case of doubt, only an invasive procedure makes it possible to make the diagnosis and to evaluate the risk for the fetus. The identification of the presence of circulating plasmatic DNA of fetal origin, as well as the emergence of the new generation sequencing technologies, allowed the development of non-invasive prenatal screening and diagnosis of various pathologies (PNID): For example, the overrepresentation of chromosome 21-specific sequences in the plasma makes it possible to demonstrate trisomy 21 (or Down syndrome) in the fetus. The entire sequencing of the maternal plasma DNA makes it possible to detect possible fetal aneuploidies.

The method of the present invention utilizes genome sequencing of maternal plasma DNA to detect CMV virus infection in the mother. Indeed, in case of active infection the virus is found in the bloodstream: viral DNA can thus be observed and quantified in the maternal plasma. This non-invasive test makes it possible to accurately determine the infective status of the woman during pregnancy, and thus improve the management of infected women by limiting the number of invasive procedures proposed.

The prenatal screening test is carried out according to the invention from maternal plasma samples. Plasma is collected after centrifugation of a blood sample from the pregnant woman and recovery of the supernatant phase. This liquid phase does not contain blood cells. Plasma circulating DNA is then extracted to allow the production of libraries that allow the sequencing of the entire genome of each sample on the Illumina HiSeql500 ™ platform, according to the manufacturer's recommendations. Sequencing is performed in this example as a single end. The raw data from sequencing as a bel file are then converted into fastq files containing all the nucleotide sequences obtained per sample. The fastq files containing the sequences are then processed according to the present invention (FIG. 9).

A total of 82 maternal plasma samples were sequenced. To this prenatal screening test were added 5 control samples (4 positive controls and 1 negative control). The sequential alignment of the sequences on the human genome and then on the modified CMV reference genome as described above (SEQ ID No. 1) makes it possible to identify, for each sample, the CMV-specific nucleotide sequences (Table 7).

Table 7. Results of the different alignment steps

The control samples passed the two quality filters of the method according to the invention, validating for these samples a sufficient depth of sequencing and alignment of the sequences on the homogeneous CMV genome. Of the positive controls analyzed, all were detected positive for CMV infection in the method according to the invention with a standardized depth greater than the threshold of 0.1424 determined by tests previously carried out on a range of samples of known viral concentration. [range: 0.2926-18.9330]. Similarly, the viral load (contaminant concentration) estimated by the method according to the invention is consistent with the theoretical viral load (theoretical contaminant concentration) of the positive control samples (R2 = 0.98). The negative control comes out well as non-infected with CMV with a standardized depth equal to 0.00019.

In the 82 primary samples tested, only 1 did not pass the quality filters because of a too low depth of sequencing: 40,450 sequences obtained against an average 19,468,992 sequences obtained for all the samples. The quantitative analysis concludes in particular to a positive infection of this sample with a standardized depth greater than 0.1424. This false-positive result highlights the need to validate the quality of sequencing by verifying that the sequencing depth is sufficient for further analysis. For the 81 remaining samples: 80 are negative for infection and 1 is positive with a viral load (contaminant concentration) estimated at 9,889 copies / mL (Figure 10).

This approach uses sequencing results used to perform other non-invasive fetal pathology screenings (fetal aneuploidy screening). The detection of viruses and other plasma microorganisms during pregnancy allows for the same cost and time to increase the number of information of interest to the health of the woman and the baby during pregnancy .

Embodiment 2: Detection and quantification of viral sequences in the diagnosis of solid tumors Oncogenesis (or malignant transformation) is the process that transforms a normal cell into a cancerous cell. This process involves the acquisition of particular properties such as uncontrolled proliferation, escape from the immune system ... (Hanahan and Weinberg, 2011). It can be induced by many events including viral infections. Indeed, since the identification by Peyton Rous of RSV virus (Rous Sarcoma Virus) capable of causing the appearance of sarcoma in chicken, many DNA and RNA viruses have been associated with the appearance of cancer in humans (Table 8). Nevertheless, the oncogenic character of the viruses has been discussed at length, in particular because of their ubiquitous nature and the time between virus infection and tumor development.

Table 8: Examples of Oncogenic Viruses

The malignant transformation of a tissue depends on the tropism of the viruses and is accompanied by the persistence of the viral genome in the tumor cells most often after integration into the DNA of the host cell (provirus). Since the integration of the virus occurs randomly in the human genome, the oncogenic action of the viruses seems to come not from an insertion mutagenesis but from a physiological modification of the cell infected by the viral proteins.

Some viruses have been directly associated with certain types of cancer: - EBV and nasopharyngeal carcinoma: Epstein-Barr virus (EBV or HSV-4) is a virus of the family Herpesviridae. It is a double-stranded DNA virus whose ubiquitous infection affects nearly 90% of the world's population. Primary infection with EBV may be asymptomatic or may be accompanied by the emergence of benign mononucleosis. EBV is a latently infected virus that persists after infection in B-lymphocytes as an episome. In healthy subjects, it is common to observe viral reactivations in the course of life.

There are several forms of latency depending on the genes whose expression persists in the host cells. A lack of control of this state can lead to the emergence of a tumor. In nasopharyngeal carcinoma (NEC), the presence of the EBV genome has been observed specifically in tumor epithelial cells. In the majority of cases, it is possible to observe the presence of multiple copies of the viral genome as an episome in infected cells (Raab-Traub, 2015). - EBV, HPV and oesophageal cancer: Esophageal cancer is one of the most common cancers observed in humans. Multiple factors are involved in the emergence of a tumor, including infectious agents such as papillomavirus (HPV), EBV or H. pylori bacteria (Xu et al., 2015). Depending on the infection causing the tumor transformation, the type of tumor that will emerge may vary. In the same way, the therapeutic approaches envisaged can be different, as well as the techniques allowing the follow-up of the disease.

The discovery of tumor DNA circulating in the plasma has made it possible to set up new strategies for detecting and monitoring non-invasive cancers (Anker et al., 1999). For example, in the case of EBV, it has been shown that the amount of virus circulating in patients' plasma was dependent on the response status of a treatment and could be used as a biomarker of anti-tumor response. A single test to quantify markers on human tumor DNA such as microsatellites, as well as viral markers, could improve the diagnosis and monitoring of the disease in a non-invasive and more accurate way. In addition, the characterization of the viral infection associated with a tumor transformation can contribute to the improvement of the diagnosis by facilitating the discrimination between several tumor types in an individual and by proposing more targeted therapeutic strategies. This is the case, for example, with oesophageal cancers that can be linked to various infections by micro-organisms. This approach is particularly useful in the case of tumors too difficult to access to allow biopsies.

The screening test according to the invention is carried out from plasma samples. Plasma DNA is extracted as previously described. The production of libraries that allow the sequencing of the entire genome of each sample on the Illumina MySeq platform is performed according to the manufacturer's recommendations. Sequencing is performed in this example as a single end. The raw data from the bel file sequencing are then converted into fastq files containing all the nucleotide sequences obtained per sample. The fastq files containing the sequences are then processed according to the invention as previously described (FIG. 9).

A total of 17 plasma test samples were sequenced and analyzed. The sequential alignment of the sequences on the human genome and then on the 9 reference genomes makes it possible to identify, for each sample, nucleotide sequences of viral origin.

All sequenced samples passed the quality tests validating both the overall sequencing quality and the quality of sequence alignment on the reference genomes studied. Of the 17 samples analyzed for 9 reference genomes (SEQ ID 1 for CMV), 3 samples were positive with a normalized depth greater than the threshold of 0.1426 previously established by the study of a range of infected samples of concentration. viral known. These 3 samples corresponding to the 3 positive controls of the study. The others were negative, as expected, for all references studied (Table 9).

Table 9: Positive results from the analysis The alignment is specific: positive controls have a depth of sequencing greater than the detection limit only for a single reference genome (Figure 11). Thus it is possible to specifically identify (specificity: 100% of 17 samples tested) and sensitive (sensitivity: 100% of 17 samples tested) samples showing viremia for one or more viral species.

In this example, the use of high throughput sequencing technologies highlights the utility of virus detection and quantification in the diagnosis and follow-up of virally induced cancer patients. The detection of low virus concentration makes it possible, in this case, to monitor and quantify, for example, the remission of patients undergoing treatment with good sensitivity. References

Adler, S.P. (2011). Screening for Cytomegalovirus during Pregnancy. Infect. Dis. Obstet. Gynecol. 2011.

Anker, P., Mulcahy, H., Chen, X.Q., and Stroun, M. (1999). Detection of circulating tumor DNA in the blood (plasma / serum) of cancer patients. Cancer Metastasis Rev. 18, 65-73.

Benoist, G., Leruez-City, M., Magny, J.F., Jacquemard, F., Salomon, L.J., and Ville, Y. (2013). Management of pregnancies with confirmed cytomegalovirus feasted infection. Feast Diagn. Ther. 33, 203-214.

Blackadar, BC (2016). Historical review of the causes of cancer. World J. Clin. Oncol. 7, 54-86.

Bolovan-Fritts, C.A., Mocarski, E.S., and Wiedeman, J.A. (1999). Peripheral blood CD14 (+) cells from healthy subjects carry a circular conformation of latent cytomegalovirus genome. Blood 93, 394-398.

Daiminger, A., Bâder, U., and Enders, G. (2005). Pre- and periconceptional primary cytomegalovirus infection: risk of vertical transmission and congenital disease. BJOG Int. J. Obstet. Gynaecol. 112, 166-172.

From Vlaminck, I., Martin, L., Kertesz, M., Patel, K., Kowarsky, M., Strehl, C., Cohen, G., Luikart, H., Neff, NF, Okamoto, J., et al. (2015). Noninvasive monitoring of infection and rejection after lung transplantation. Proc. Natl. Acad. Sci. U.S.A. 112, 13336-13341.

Dunn, W., Chou, C., Li, H., Hai, R., Patterson, D., Stolc, V., Zhu, H., and Liu, F. (2003). Functional profiling of a human cytomegalovirus genome. Proc. Natl. Acad. Sci. U.S.A. 100, 14223-14228.

Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674.

James, S.H., and Kimberlin, D.W. (2016). Advances in the prevention and treatment of congenital cytomegalovirus infection. Curr. Opin. Pediatr. 28, 81-85.

Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009). Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25.

Liesnard, C., Donner, C., Brancart, F., Gosselin, F., Delforge, M.L., and Rodesch, F. (2000). Prenatal diagnosis of congenital cytomegalovirus infection: prospective study of 237 pregnancies at risk. Obstet. Gynecol. 95, 881-888. NIH HMP Working Group, Peterson, J., Garges, S., Giovanni, M., Mclnnes, P., Wang, L., Schloss, J.A., Bonazzi, V., McEwen, J.E., Wetterstrand, K.A., et al. (2009). The NIH Human Microbiome Project. Genome Res. 19, 2317-2323.

Pereira, L., Maidji, E., McDonagh, S., and Tabata, T. (2005). Insights into viral transmission at the uterine-placental interface. Trends Microbiol. 13, 164-174.

Raab-Traub, N. (2015). Nasopharyngeal Carcinoma: An Evolving Role for the Epstein-Barr Virus. Curr. Top. Microbiol. Immunol. 390, 339-363.

Xu, W., Liu, Z., Bao, Q., and Qian, Z. (2015). Viruses, Other Pathogenic Microorganisms and Esophageal Cancer. Gastrointest. Tumors 2, 2-13.

Claims (1)

Claims 1 - Method for determining the presence and quantification of at least one microorganism in a human biological sample comprising total nucleic acids, comprising the following steps: - extraction of the total nucleic acids from said biological sample - high-throughput sequencing of said acids total nuclei - computer processing of sequencing data consisting in counting the total number of sequences or filtering said sequences as a function of a quality score; filtering human sequences by sequence alignment with the sequences of the human referenced genome characterized in that said computer processing steps further comprise: an initial calibration step of characterizing a plurality of reference samples whose concentration of microorganism is known, and recording a matrix corresponding to the depth determined by said actinisation, and the corresponding known microorganism concentration - a 0-5% masking step of the nucleotides of the reference genome of at least one microorganism relative to the total nucleotides of said reference genome, said masked nucleotides corresponding to sequences of low complexity - filtering the sequences of at least one microorganism by alignment of non-human sequences with reference genome sequences of at least one microorganism of at least one reference sample - the calculation of the sequencing depth of the reference genome of at least one microorganism - the validation of the homogeneity of the genome coverage of at least one microorganism consisting of calculating the standard deviation of the genome sequencing depth at the reference of said genome of at least one microorganism - the determination of an indicator of the quantification of at least one microorganism function of said pro sequencing melter of the reference genome 2 - A method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 1 characterized in that the step of calculating the depth of sequencing of the genome of reference of at least one microorganism comprises a depth normalization treatment with respect to the total number of sequences resulting from the sequencing. 3 - Equipment for the determination of the presence and quantification of at least one microorganism in a biological sample characterized in that it comprises a computer controlled by a computer program for carrying out treatments of: - initial calibration consisting of characterizing a plurality of reference samples whose concentration in microorganism is known, and recording a matrix corresponding to the depth determined by said characterization, and the corresponding known microorganism concentration - masking of 0 to 5% of the nucleotides of the reference genome of at least one microorganism relative to the total nucleotides of said reference genome, said masked nucleotides corresponding to sequences of low complexity - filtering the sequences of at least one microorganism by non-sequence alignment -humans with the sequences of at least one reference genome of said microorganism of at least one reference sample - calculation of the sequencing depth of said reference genome - validation of the homogeneity of the genome coverage of at least one micro-organism consisting in calculating the standard deviation of the depth of sequencing of the genome to the reference of said genome of at least one microorganism - determination of a microorganism concentration indicator according to said depth of sequencing of the reference genome. 4 - Computer program controlling a calculator of an equipment for the determination of the presence and quantification of at least one microorganism in a biological sample characterized in that it controls the carrying out of the treatments of: - initial calibration consisting of characterizing a plurality of reference samples whose concentration in microorganism is known, and recording a matrix corresponding to the depth determined by said characterization, and the corresponding known microorganism concentration - masking of 0 to 5% of the nucleotides of the reference genome of at least one microorganism relative to the total nucleotides of said reference genome, said masked nucleotides corresponding to sequences of low complexity - filtering the sequences of at least one microorganism by non-sequence alignment -humans with reference genome sequences of said microorganism of at least one n reference sample - calculation of the sequencing depth of the reference genome - the validation of the homogeneity of the genome coverage of at least one microorganism consisting in calculating the standard deviation of the depth of sequencing of the genome reference to said genome of at least one microorganism - determination of a microorganism concentration indicator according to said depth of sequencing of the reference genome. Kit for the determination of the presence and quantification of at least one microorganism in a biological sample according to the process according to claim 1, characterized in that it comprises: a plurality of tubes each containing a dilution of least one DNA sequence of a reference genome of at least one microorganism in a control DNA - a tube containing DNA controls - the nucleotide sequence of the reference genome of at least one microorganism of which 0 to 5% of the nucleotides are masked, said masked nucleotides corresponding to sequences of low complexity. 6 - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 1 for determining the concentration of a parasite. 7 - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 6 characterized in that said parasite is a prokaryotic organism. 8 - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 6 characterized in that said parasite is a virus. 9 - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 6, characterized in that said parasite is a eukaryotic organism. - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 6 for the prenatal diagnosis characterized in that said parasite is a cytomegalovirus. 11 - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 6 for the diagnosis of a cancer induced by the Epstein Barr virus and / or the monitoring of a treatment anticancer agent characterized in that said parasite is an Epstein Barr virus. 12 - Application of the method for determining the presence and quantification of at least one microorganism in a biological sample according to claim 6 for monitoring a transplant characterized in that said parasite is a cytomegalovirus virus. 13 - Kit for prenatal diagnosis implementing the method according to claim 1 characterized in that said reference genome sequence of at least one microorganism is a cytomegalovirus of which up to 0.06% of the nucleotides are masked. 14 - Kit for monitoring a transplant implementing the method according to claim 1 characterized in that said reference genome sequence of at least one microorganism is a cytomegalovirus. 15 - Kit for the diagnosis of a cancer induced by the Epstein Barr virus and / or the monitoring of an anticancer treatment implementing the method according to claim 1 characterized in that said reference genome sequences of at least a microorganism is an Epstein Barr virus.