Method for detecting the presence of a biological species of interest by iterative real-time sequencing.

Abstract

The invention relates to a metagenomic analysis of a sample for the purpose of detecting the presence of a species of interest within the sample. The analysis is performed iteratively. During each iteration, sequences corresponding to each species of interest are identified and counted. The iterations cease when the presence of a species of interest is confirmed or when a maximum number of iterations have been performed. The detection of a species of interest can be followed by a more precise characterization of the genome of said species of interest. The characterization involves additional iterations. Figure 1.

Description

Title of the invention: Method for detecting the presence of a biological species of interest by iterative real-time sequencing. Technical field

[0001] The technical field of the invention is the identification of a biological species of interest by metagenomic analysis, by implementing a real-time sequencing method. PREVIOUS ART

[0002] Recently, so-called "real-time" sequencing techniques have emerged. These platforms allow immediate or slightly delayed access to sequences as they are read. In addition, the quantity of sequences read is no longer limited by the need to mobilize the nucleic acids onto a support, which potentially allows access to all the nucleic acid sequences present in the sample analyzed. An example of a method is nanopore DNA sequencing, as implemented by the MinlON and GridION sequencing platforms from Oxford Nanopore Technologies.

[0003] Nanopore DNA sequencing is based on the passage of a molecule, comprising a strand of oligonucleotides, through a nanopore, the latter forming a channel. When the molecule passes through the channel, a potential difference can be measured, on either side of the channel, which depends on the nature of each base forming the strand. The potential difference allows a distinction between the five standard bases of DNA or RNA (G, C, T or U). Thus, during passage through the nanopore, the strand of oligonucleotides induces a temporal sequence of potential differences, from which the sequence of bases forming the strand is determined.

[0004] Document WO2021 / 013900 describes a method for detecting a species of interest, in particular a bacterium, in a sample. A control species has been previously added to the sample, in a known quantity. After sequencing, the number of sequences corresponding to the control species is used to quantify a concentration of the species of interest or to determine a minimum detectable concentration of the species of interest in the sample. The use of the control species makes it possible to ensure that the extraction process and the possible amplification of the nucleotide sequences is carried out correctly. Since the quantity of control species introduced into the sample is known, it can be used to quantify a concentration or a minimum detectable concentration of the species of interest.

[0005] The inventor has adapted the method previously described to the use of real-time sequencers delivering their digital nucleotide sequences (also called "reads" and corresponding to DNA or RNA sequences) as the nucleotide fragments in the sample are read. Indeed, these allow an analysis result to be obtained more quickly, for example a few hours or even a few tens of minutes. Statement of the invention

[0006] A first subject of the invention is a method for detecting a biological species of interest potentially present in a sample, the biological species of interest having a known or partially known genome, the sample comprising a mixture of different biological species, the method comprising the steps:

[0007] a) addition of a control species to the sample, the control species having a known genome, the control species being added at a known concentration in the sample; b) extraction of nucleic acids from the sample; c) sequencing a portion of the nucleic acid sequences extracted during step b) using a real-time sequencer for a predetermined duration or until a predetermined number of sequencing operations is obtained; (d) following step (c), assignment of sequences resulting from step (c) to the species of interest as well as to the control species; (e) updating the quantities of sequences respectively associated with the species of interest and the control species, such that the quantities of sequences associated with the control species and the species of interest are: during a first iteration of step e), the quantities of sequences assigned to the control species and to the species of interest during step d) added to initial quantities of sequences; during each reiteration of step e), the quantities of sequences (n\O], n'spc^ assigned to the control species and to the species of interest during step d) added to the quantities of sequences associated with the control species and to the species of interest resulting from the previous iteration; fjcomparison of the quantity of sequences associated with the species of interest updated in step e) with a threshold of the number of sequences;

[0008] and / or

[0009] comparison of the quantity of sequences associated with the updated control species

[0010] in step e) with a control threshold; - g) based on at least one comparison made during step f), detection of the presence of the biological species of interest in the sample, or moving on to step h); - h) repetition of steps c) to g) until an iteration stopping criterion is reached.

[0011] According to one embodiment, step d) of each iteration is carried out for a predetermined duration or until a predetermined number of sequences have been sequencing.

[0012] The method may be such that:

[0013] - during step f) of an iteration, when the quantity of sequences associated with the species of interest is greater than the threshold of the number of sequences; - during step e) of the iteration, when the quantity of sequences associated with the control species is strictly greater than zero;

[0014] step g) includes an estimation of a concentration of the species of interest as a function of:

[0015] - the quantity of sequences associated with the species of interest during the last update (i.e. during the last step e); - the quantity of sequences associated with the control species during the last update; - the added concentration of the control species.

[0016] The method may be such that:

[0017] - during step f) of an iteration, when the quantity of sequences associated with the species of interest is greater than the threshold of the number of sequences; - during step e) of the iteration, when the quantity of sequences associated with the control species is zero;

[0018] during step g), the concentration of the species of interest is considered to be higher than the concentration of the control species.

[0019] The method may be such that:

[0020] - during step f) of an iteration, when the quantity of sequences associated with the species of interest is less than the threshold of the number of sequences; - during step f) of the iteration, when the quantity of sequences associated with the control species is greater than the control threshold;

[0021] during step g), if the quantity of sequences associated with the species of interest is zero, the concentration of the species of interest is considered to be zero.

[0022] The method may be such that:

[0023] - during step f) of an iteration, when the quantity of sequences associated with the species of interest is less than the threshold of the number of sequences; - during step f) of the iteration, when the quantity of sequences associated with the control species is greater than the control threshold;

[0024] during step g), if the quantity of sequences associated with the species of interest is non-zero, step g) includes an estimation of a concentration of the species of interest as a function of:

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035] the quantity of sequences associated with the species of interest; of the quantity of sequences associated with the control species; of the added concentration of the control species. The concentration of the species of interest can be estimated from a ratio: self SPC where: Lspc ^soi SPC - Lspc and £yo / are the genome lengths of the control species and the species of interest respectively; - Ct ^SPC S°nt resPec^vement 'cs quantities of sequences respectively associated with the species of interest and the control species resulting from the last update; - Cspc is the concentration of the control species added to the sample. The process can be such that, following the achievement of the iteration stopping criterion, - if the quantity of sequences associated with the species of interest is lower than the threshold of the number of sequences; - and if the quantity of sequences associated with the control species is lower than the control threshold; - the process generates information according to which neither the presence nor the absence of the species of interest can be confirmed. According to one possibility, the method comprises taking into account a decision threshold, the method aiming to compare a concentration of the species of interest with the decision threshold, the method being such that during step a), the concentration of the control species is between 0.01 times and 100 times the decision threshold. According to one possibility, in step h), the iteration stopping criterion is reached when: - the quantity of sequences associated with the species of interest exceeds the sequence number threshold; - and / or the quantity of sequences associated with the control species exceeds the control threshold; - and / or the cumulative duration of steps c) reaches a predetermined maximum duration; - and / or the number of cumulative sequencings during steps c) reaches a predetermined maximum number; - and / or the number of iterations reaches a predetermined maximum number of iterations. According to one embodiment, following detection of a presence of the species of interest in the sample, the process involves the steps:

[0036] - i) sequencing a portion of the nucleotide sequences extracted during step b) using the real-time sequencer for a predetermined duration or until a predetermined number of sequencing operations is obtained; - j) following step i), assignment of sequences resulting from step i) to the detected species of interest; - k) updating the quantities of sequences associated with the species of interest, such that the quantities of sequences associated with the species of interest are: • during a first iteration of step k), the quantity of sequences assigned to the species of interest detected during step j) added to the quantity of sequences resulting from the last iteration of steps c) to g); • during each repetition of step k), the quantity of sequences assigned to the species of interest during step j) added to the quantity of sequences resulting from the previous iteration of step k); - 1) repetition of steps i) to k) until a stopping criterion is reached of iterations.

[0037] Steps i) to k) can be repeated until a predetermined number of iterations is reached.

[0038] According to one possibility:

[0039] - during each iteration of steps i) to k), following step k), the method comprises a determination of a sequencing depth of the biological species of interest, the sequencing depth corresponding to a ratio between the cumulative length of the sequences associated with the species and the length of the genome of the species; - steps i) to k) are repeated until a predetermined sequencing depth is reached.

[0040] According to one embodiment, following the stopping of the iterations of steps i) to k), a detection of a typical sequence of the genome of the species of interest, the typical sequence comprising an antibiotic resistance marker or a virulence marker of the species of interest.

[0041] Preferably, the sequencer is configured to carry out successive sequencing of different nucleic acid sequences, one after the other, the sequencing of each sequence being considered to be carried out in real time.

[0042] A second object of the invention is a device for metagenomic analysis of a sample, comprising:

[0043] - a fluidic chamber, intended to receive the sample; - a sequencer configured to perform real-time sequencing of the sample; - a processing unit, comprising a bioinformatics module, programmed to assign sequences resulting from the sequencer to a species, and a control module, programmed to implement steps c) to h) of a method according to the first subject of the invention.

[0044] A third subject of the invention is a recording medium, readable by computer, or downloadable, comprising the instructions for implementing steps c) to h) of a method according to the first subject of the invention.

[0045] The invention will be better understood upon reading the description of the exemplary embodiments presented in the remainder of the description, in conjunction with the figures listed below. FIGURES

[0046] Figures 1 and 2 diagram the main steps of a method according to the invention.

[0047] [Fig.l] concerns a phase of detection of a species of interest.

[0048] [Fig.2] concerns a characterization phase of each species of interest detected during the detection phase.

[0049] [Fig. 3] shows schematically a device configured to implement the invention. DISCLOSURE OF PARTICULAR EMBODIMENTS

[0050] The objective of the method is to be able to detect the presence of a biological species of interest SOI in a sample. The acronym SOI stands for "Species of Interest". In the event of detection, the method can allow absolute quantification of the species of interest SOI, so as to allow comparison with a decision threshold SD.

[0051] By biological species is meant a microorganism, for example a bacterium, or a virus, a fungus, an archaeobacterium, an amoeba, a protist, a microalgae. A biological species can also be a cell or any other material or entity comprising a sequencing nucleic acid.

[0052] When the sample comes from a human or animal organism, the biological species of interest may be a pathogenic species or a species identified as having an impact on the health or functioning of the human or animal body, such as, for example, intestinal bacterial species, or bacteria impacting the effectiveness of an anti-cancer treatment. When the sample comes from a sample taken in an industrial process or in the environment, the biological species of interest may be a species considered to be a contaminant, or a species of interest having importance in an industrial process or in the environment, and whose presence or concentration is to be controlled.

[0053] The species of interest has a known, or partially known, genome. The genome, or its known portion, is made up of sequences, called sequences of interest.

[0054] The method can simultaneously address several species of interest. Also, the term a species of interest is to be interpreted as meaning at least one species of interest.

[0055] The decision threshold SD is a threshold making it possible to characterize a load of the biological species of interest, for example of a microorganism, depending on the intended application. It is for example established from a regulatory, or health or industrial limit. For example, when the application is used in clinical diagnosis assistance, the biological species of interest being a bacterium, the decision threshold may be a concentration below which the presence of the bacterium corresponds to a normal presence, that is to say a non-pathological development, and beyond which the presence of the bacterium is considered pathological, corresponding for example to an infection.When the invention is applied in an industrial process, the decision threshold may correspond to a conformity value, such that above the decision threshold, the sample is considered non-conforming, and below the decision threshold, the sample is considered compliant. Whatever the application, when the concentration of the biological species of interest is greater than or equal to the decision threshold, it is defined as critical. In certain applications, for example in the manufacture of products undergoing fermentation, a concentration of biological species of interest may be considered critical if it is lower than a decision threshold, the latter corresponding to a minimum admissible concentration of the biological species.

[0056] The sample is generally a sample taken from the environment, or from an organism, dead or alive, or from an agri-food or manufactured product. The sample may also have been taken from an industrial facility for process control purposes. Also, the sample contains different biological species, not having the same genome. In particular, when the sample results from a sample taken from an organism, for example a human or animal organism, the sample contains a significant quantity, or even the majority, of cells originating from the organism from which the sample was taken. The genomes of human or animal organisms are 1,000 to 100,000 times larger than the genomes of prokaryotic organisms. In addition, the sample generally contains biological species naturally present in the sample, and not likely to cause pathology or critical contamination.For example, when the sample is a bronchoalveolar sample, it contains bacterial flora naturally present in the lungs. When the sample is a stool sample, it contains bacterial flora naturally present in the digestive tract. In this respect, when the biological species of interest is a bacterium or a virus, the nucleic acids from the biological species of interest may be in the minority in the sample.

[0057] The sample contains so-called "matrix" species, endogenous in the sample, and likely to mask metagenomic information relating to the biological species of interest. For example, when the sample is taken from yogurt, meat or a vaccine, it contains matrix species representative of these environments. In the case of a sample from an organism, the matrix contains the cells making up the organism.

[0058] An important aspect of the invention is that the presence of the species of interest in the sample is detected by implementing a real-time sequencer mentioned in the prior art. Such a sequencer makes it possible to obtain exploitable data, relating to the sequences present in the sample, more quickly than prior sequencers. The real-time sequencer is configured to successively collect signals, representative of successive detections of bases of the same sequence. Thus, a real-time sequencer is configured to generate a sequence in real time, as it is decoded. Sequencing of different sequences present in the sample is carried out successively, each sequence being sequenced one after the other. This does not exclude the fact that certain sequences can be the subject of simultaneous sequencing, in parallel.For example, when the sequencer is a nanopore sequencer, each nanopore "reads" several different sequences in parallel. Each nanopore can read several different sequences successively.

[0059] Another advantage of such a sequencer is the possibility of performing real-time analysis of the sequences and the quantity of sequences read is relative to the duration of the sequencing. Depending on the available sequencing results, sequencing can be continued or stopped, which saves time and provides great flexibility of use. The sequencer can be of the nanopore type, as described in the prior art. More generally, the invention relates to the use of a sequencer configured to generate, in real time, sequences, the latter being identified following their sequencing. Thus, the sequencer is configured to sequentially generate a list of identified sequences, the latter being designated by the term "reads". The lists of reads are generated at regular intervals, for example every n minutes, l being for example equal to 10. It is also possible to configure the quantity of sequences read in each list.The number of sequences (or reads) in each list can for example be 4000.

[0060] As in application WO2021 / 013900, one of the objectives of the invention is to evaluate the extent to which a metagenomic analysis is usable. This involves in particular evaluating the conformity of all the steps from the preparation of the sample, excluding sampling, to the bioinformatic analysis of the sequencing data. To this end, a control species, noted SPC, acronym for Sample Processing Control, is added to the sample. A function of the control species is to enable monitoring of the proper conduct of the nucleic acid extraction and sequencing steps, described below. The control species SPC may be a known biological species, the genome of which is also known, preferably in its entirety. The control species SPC may be a natural biological species. It may also be an artificial species, for example an encapsidated RNA (ribonucleic acid). Preferably, the control species SPC is not initially present in the sample taken, or in a negligible amount. Preferably, the content of control species SPC initially present in the sample, i.e. present before the addition, is preferably at least 10 times lower, or preferably at least 100 or 1,000 times lower than the added concentration CSPc of the control species SPC in the sample. The control species SPC may, for example, be a bacterium.It is important that the CSPc concentration of the added SPC control species is controlled.

[0061] The control species can be chosen taking into account the aspects listed below:

[0062] - The control species must preferably be distinguished from the organisms naturally present in the sample, or endogenous organisms, as well as from the species of interest sought: thus, the bioinformatics tool can precisely identify the sequences resulting from the sequencing of the SPC. - The quantity of sequences assigned to the control species during sequencing must be sufficient to be detected correctly, without masking the useful information corresponding to the sequences of the biological species of interest. In other words, the control species is preferably detectable by high-throughput sequencing, not everything being preponderant in the sample. In particular, when one wishes to determine a positivity (concentration of the species above the decision threshold) or a negativity (concentration of the species below the decision threshold), it is preferable that the control species presents at least one of the following characteristics: • The size of its genome is preferably similar, or at least comparable, to the size of the genome of the biological species of interest. More particularly, the size of the genome of the control species is between 0.1 times and 10 times the size of the genome of the biological species of interest. • The CSPc concentration of the control species can be determined as a function of the decision threshold. The CSPc concentration of the added SPC control species can be, for example, between 0.001 times and 1000 times, and preferably between 0.01 and 100 times the decision threshold. • Nucleic acids of the control species SPC undergo a similar treatment to the nucleic acids of the species of interest during the sample preparation, extraction and sequencing steps, and preferably: • the percentage of GC bases (Guanine, Cytosine) is preferably close to the percentage of GC bases of the biological species of interest; By close to, we mean between 75% and 125%, and preferably between 80% and 120%. • the control biological species preferably comprises, when the biological species of interest is a bacterium, an intact membrane or cell wall or, when the biological species of interest is a virus, a protein envelope. This condition also allows monitoring of the lysis steps or the extraction of nucleic acids from the biological species of interest. - The nucleotide sequences of the control species preferably do not contain genomic markers, such as antibiotic resistance markers or virulence markers, so as not to distort the results of a possible antibiotic sensitivity test by the presence of such markers in the genome of the biological species of interest. Preferably, the nucleotide sequences of the control species do not contain any other gene of clinical or industrial interest whose presence is likely to be controlled. - The control species is preferably easily manipulated, in particular: • being harmless to humans or the environment; • and / or being resistant to heat treatments such as lyophilization lization or freezing, which makes storage easier. - The control species must not form spores, or only marginally. - The control species must have a sensitivity to lysis close to that of the biological species of interest. - The control species may be present in the form of beads, each bead containing a calibrated concentration of control biological species in lyophilized form.

[0063] It is specified that a single species of SPC control can be used, or that several species of control, of different types, can be used.

[0064] The method firstly comprises a phase of detecting a species of interest, the steps of which are now linked to [Fig.l].

[0065] Step 10: sample collection.

[0066] In this example, the sample is taken from a living human organism, at diagnostic aid purposes. However, the invention is not limited to an application in the field of living organisms. The sample can be taken from a natural, industrial or hospital environment, in order to verify compliance with a decision threshold.

[0067] Step 20: Adding the control species.

[0068] The added concentration CSPc of the control species SPC is preferably known precisely, for example with the same precision as that desired for the quantification of the species(s) of interest. Indeed, it can make it possible, provided that certain conditions are met, to quantify the concentration of biological species of interest in the sample, the control species then forming a calibrator. The term added concentration designates the concentration of the control species in the sample due to the addition of the control species.

[0069] In the description of steps 30 to 60, the addition of a single type of control species to the sample is used as an advantageous example. The control species then fulfills the quality control function of the steps of the metagenomic analysis, as well as the function of calibrator, allowing possible quantification of the concentration of the biological species of interest.

[0070] At the end of step 20, an added concentration CSpc of the control species in the sample is available. The added concentration CSpc of the control species is preferably expressed in GEq / mL (genome equivalent per mL).

[0071] It is noted that clinical detection thresholds in bacteriology (eg the threshold beyond which an infection is diagnosed) are generally expressed in CFU / mL (colony forming unit), however the proposed quantification method allows quantification of nucleic acids without information on genome equivalence. It is therefore recommended to determine, for the SPC, the correlation between the number of colony forming units by culture for example on an agar medium and the equivalence in number of genomes for example by quantitative PCR. For example, it has been established that in a bioball® B. subtilis strain ATCC 19659 BioBall® MultiShot 10E8 (bioMérieux, Catalog #416721) a colony forming unit was equivalent to a genome.

[0072] The added concentration may be defined as a function of the decision threshold SD. Preferably, the added concentration CSPc of the control species is equal to the decision threshold SD, or close to the decision threshold SD, for example within ± 50%. Thus, 0.1 SD < CSpc < 10 SD. The impact of the added concentration CSpc of the control species is described below.

[0073] Step 30: lysis and extraction of nucleic acids.

[0074] During this step, the cells of the sample, and in particular the cells of the biological species of interest and the control species, undergo lysis, for allow extraction of their DNA. Different strategies can be envisaged:

[0075] • lysis can be configured to preferentially target the biological species of interest. According to such a scenario, the control species must have the same sensitivity to lysis as the biological species of interest, or a sensitivity to lysis considered equivalent. • the lysis may include a first lysis, intended to lyse essentially cells other than the species of interest. Such a first lysis may for example be envisaged when the biological species of interest is very much in the minority compared to the cells of a matrix making up the sample. Following the first lysis, the released nucleic acids are removed, then a second lysis is carried out, targeting the biological species of interest. According to such a scenario, the control species is preferably resistant to the first lysis, and not resistant to the second lysis.

[0076] Following lysis, the DNA is extracted from the sample, for example according to the extraction method described in WO2014 / 114896. Step 40: Sequencing

[0077] The DNA extracted from the sample undergoes sequencing using a sequencer allowing real-time access to the sequences read, for example with nanopores. This may for example be a MinlON ® type sequencer from Oxford Nanopore Technologies. The sequencer is connected to a dedicated fluidic chamber, comprising the nanopores. The sample is introduced into the fluidic chamber. The sequencer is connected to a processing unit allowing the analysis of the identified sequences. The processing unit comprises a microprocessor, configured to execute instructions from bioinformatics software, to identify the sequences analyzed. The processing unit is for example a MinIT type computer, comprising the computer environment dedicated to sequencing.

[0078] The invention applies particularly to so-called “real-time” sequencers having the following characteristics:

[0079] - a capacity to read all the sequences present in the sample unlike sequencers based on immobilization of the sequences on a support for their reading, these sequencers therefore sequencing a predefined maximum number of sequences corresponding to the number of fixing elements of the nucleotide sequences. This implies in particular that the number of sequences produced is proportional to the sequencing duration. - an ability to output raw data (e.g. reads or raw signals such as electrical potentials) as they are collected, or to output raw data in batches at a frequency defined by batch of sequences of a given size or regularly at a frequency fixed (for example, a batch of raw data per minute or per ten minutes).

[0080] A particularity of “real-time” sequencing is that the results of reading the DNA or RNA sequences are available in real time at the output of the sequencer, or in near real time. By near real time, we mean that the reads (or the raw signals measured such as the electrical potentials) are produced at the output of the sequencer in batches after a time much shorter than the total time required to sequencing the acids present in the sample or much shorter than the total time required to consume the reagents required for sequencing. Each batch can thus be produced in a few minutes or a few tens of minutes. The real-time sequencer has the capacity to produce sequences (reads) in real time. However, we generally wait until a certain number of sequences have been decoded, in a file generated at regular intervals. For example, the Oxford Nanopore MinlON sequencer generates, by default, in near real time files of the type “.fast5 » in batches of 4,000 potential difference sequences that can be converted in near real-time by a computer into a .fastq file in batches of 4,000 reads. The .fast5 extension denotes raw data files. The .fastq extension denotes files storing sequences, as well as associated quality scores. This allows interactive implementation of the process, so that sequencing can be stopped or continued depending on the results obtained.

[0081] The sequencing may in particular be carried out iteratively, each iteration corresponding to a sequencing of a predetermined number of sequences (or batch) or to a sequencing carried out during a predetermined duration. During each iteration, following the sequencing, sequences are potentially assigned to the control species SPC or to the biological species of interest SOI, or to each biological species of interest or to other species present in the sample.

[0082] Each iteration corresponds to a rank i. i is an integer between 1 and / , / corresponding to the maximum number of iterations described in connection with step 70, or to a maximum number of sequences (reads) obtained. During each iteration, the sequencing assigns numbers of sequences n'SOI and n^pç respectively to the species of interest SOI and to the control species SPC.

[0083] The following steps are described in relation to one species of interest SOI. They can be implemented with several different species of interest. Step 50: Updating the number of sequences.

[0084] During this step, the numbers of sequences respectively assigned, during step 40, to each biological species of interest and to the control species are respectively the number of sequences respectively associated with the species

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093] biological species of interest SOI and the control species SPC during the previous iteration. This step corresponds to an update of the number of sequences respectively associated with the species of interest SOI and the control species SPC, according to the expressions: N SOI = NSOI + n'SOI (1) spc = NSPC + nlSPC (2) where n\O] and n'SPC are the numbers of sequences respectively assigned to the species of interest and the control species during step 40 of iteration i. The notation n1 corresponds to a number of sequences assigned to a species during iteration i, while the notation corresponds to a total number of sequences associated with a species for all iterations up to iteration i. During the first iteration (' = 1), we consider initial values ​​and N^PC predetermined, generally zero: N® — N^PC = 0 According to one possibility, the numbers of sequences N'SO[ or NlSPC Can be nor weighted by a reference quantity. The reference quantity can for example be a total quantity of sequences produced during sequencing, up to the current iteration i. Step 60: Comparisons to thresholds During this stage, the number of sequences associated with each species of interest SOI is compared to a threshold of the number of Ssoh sequences which corresponds to the minimum number of reads associated with the species of interest SOI required to stop sequencing and begin quantitative interpretation of the presence of the species of interest in the sample. The sequence count threshold corresponds to a threshold of the number of sequences, or threshold of the number of reads, associated with the species of interest. The number N* of sequences associated with the control species SPC is compared to a control threshold $spc- which corresponds to the minimum number of reads associated with the SPC to stop sequencing when the number of reads associated with each of the SOIs does not reach the threshold of the number of sequences SSoi- - When xd < 9 and Ad < ç: cf. step 61; / V SOI to SOI SPC s ^SPC r When and > 0: see step 62; When and 2^ = 0: cf- step 63 - When xd < ç and xd x ç: cf. step 64 SOI x to SOI SPC - ÛSPC 1 The determination of the sequence number threshold and the control threshold are described below. Step 61

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104] During this step, no conclusion can be drawn regarding the presence of the SOI species of interest in the sample. A new iteration of steps 40 to 60 is necessary, unless an iteration stopping criterion is reached, as described in connection with step 70 Step 62: During this step, it is possible to conclude the presence of the species of interest SOI in the sample and to quantify its concentration, for example by the expression: N1 T / \ C _ self x LSpc c U ■SOI - X Tf XX a PP* SPC \ / Or : - LSpc and Po / are the genome lengths of the control species and the biological species of interest, respectively. - a is a correction factor determined empirically, based on training samples whose concentration of the biological species of interest is known. The correction factor a allows for differences in the efficiency of the sequencing process of the biological species of interest and the control species to be taken into account. By default, a can be considered = 1. This unit value allows for sufficient absolute quantification to be obtained to determine the positivity or negativity of a sample in relation to the decision threshold. When the added concentration is expressed in GEq / mL, the concentration of the biological species of interest is also expressed in the same unit. Step 63 During this step, the species of interest is considered to be present in the sample at a concentration higher than the concentration of the control species, but it is not possible to determine its absolute concentration. Csoi GSPC (4) Step 64 During this step, if — 0' it is possible to conclude in the absence of the species of interest SOI in the sample. Otherwise it is possible to quantify its concentration, with the formula described in step 62. Step 70: Steps 40 to 60 are repeated until a criterion for stopping the iterations is reached. Preferably, each iteration i corresponds to a predetermined sequencing duration, and preferably identical for each iteration, or to a predetermined number of sequences read (number of reads). The iteration stopping criterion is considered to be met when the conditions leading to steps 62 or 63 or 64 are met.

[0105] The iteration stopping criterion may also be a number of iterations equal to a maximum number of iterations / , the latter having been predetermined. For example, the iteration stopping criterion is considered to be reached when the sequencing duration is greater than or equal to a predetermined duration, for example 12 hours, or when the total number of sequences from the sequencer is greater than a predetermined maximum number.

[0106] Step 80 Alert.

[0107] Step 80 is performed if m1 < ç and m1 < ç , i.e. following the 1 ivsol s jsol ivSPC^ toSPC last iteration of rank / , the numbers of sequences respectively associated with the species of interest and the control are respectively lower than the threshold of the number of sequences and the control threshold. In this case, an alert signal is issued, indicating that the presence or absence of the species of interest in the sample cannot be confirmed.

[0108] Step 90 Stop

[0109] Following step 70 or a possible step 80, the method can be stopped.

[0110] According to one possibility, several species of interest, different from each other, are analyzed simultaneously. When one of the conditions indicated in steps 62, 63 and 64 is encountered, the sequencing can be stopped and each species of interest is interpreted according to these criteria.

[0111] Optionally, after the presence of a species of interest in the sample has been reported, following steps 62 or 63 or 64, it is possible to have additional genomic information regarding the identified species of interest. This may in particular be the presence of a specific marker in the genome, for example an antibiotic resistance marker or a virulence marker. The antibiotic resistance or virulence markers are known.

[0112] The iterative steps 100 to 130, described below, in connection with FIG. 2, can be implemented to confirm the presence of the specific marker in the genome of the identified species of interest. If i' designates the rank of the iteration during which the iteration stopping criterion of step 70 was reached, the steps 100 to 130 are implemented for / < / < J / corresponds to the rank of the last iteration.

[0113] Step 100

[0114] This step is similar to step 40. Only the sequences corresponding to each species of interest SOI detected following the iterations of steps 40 to 80 are counted. Step 110

[0115] This step corresponds to step 50, except that we proceed to an update of the number of sequences of each species of interest according to the expression: 101161 N'so, = N's<'„ + n'SOI (1)

[0117] Step 120: repeat steps 100 to 110, unless an iteration stopping criterion is reached. The iteration stopping criterion may be the reaching of a maximum number of iterations / . It is recalled that the latter may correspond to a maximum sequencing duration or to a maximum number of sequences identified.

[0118] According to one possibility, a sequencing depth of the species of interest SOI is determined during each iteration. The sequencing depth corresponds to a ratio between the cumulative length of the sequences associated with the species of interest and the length of the genome of the species of interest. When the sequencing depth reaches a threshold depth, the criterion for stopping the iterations is reached. The iterations are stopped.

[0119] Step 130: Characterization

[0120] During this step, the sequences corresponding to the desired marker are identified and counted.

[0121] Determination of the threshold for the number of Ssoi sequences and the Sspc control threshold

[0122] The implementation of the method previously described presupposes a prior determination:

[0123] - the threshold of the number of sequences Ssoh to which the number of sequences of the species of interest ^O} is compared during each step 60; - and the Sspc control threshold, at which the number of control sequences is compared at each step 60.

[0124] The sequencing times required to produce numbers of reads corresponding to the species of interest SOI were estimated to be respectively 1, 10, 100, 1,000 and 10,000, taking into account different CSOI concentrations of the species of interest SOI in the sample.

[0125] The sequencing duration TTR (Time to Resuit) is estimated according to the expression:

[0126] ~ Sso} x CADNtot x T f astq ^sample / c\ f astq \ /

[0127] where:

[0128] - Ssoi is the threshold of the number of sequences; - CADNtot is the total DNA concentration in the sample: this includes the concentrations of the species of interest, the control species and the DNA of no interest (human DNA and commensal flora); - & fastq is a quantity of reads obtained during each iteration, i.e. in each generated .fastq file. For example, Rfastq = 4000. - Tfastq is an average sequencing time to obtain a .fastq file with a quantity of reads equal to Rfastq; - sample cst 'c number of samples sequenced simultaneously. Nsa le corresponds to a proportion of sequenced DNA that can be attributed to the sample. Nsa ie is different from 1 when different samples are sequenced simultaneously. In this case, Nsa le denotes the total amount of sequenced DNA divided by the amount of DNA in the sample considered.

[0129] Table 1 gives the sequencing time (in HH:MM) which is calculated as a function of the threshold of the number of sequences Ssoh and the concentration CSOJ of species of interest SOI expressed as a function of the decision threshold SD. It was considered that the control species was added according to a concentration CSPc=SD and that the sum of human DNA and commensal flora was equivalent to 100SD. The total amount of DNA is then equal to CADNtat = 101572 + CSOJ.The sample is considered to be sequenced at the same time as three other samples (Nsampie=4), each of the samples being sequenced with an equivalent amount of DNA with a sequencer that generates read files with Rfastq = 4,000 reads with a frequency of approximately Tfastq = 3 minutes.

[0130] [Tables 1] G SELF \S self 1 10 100 1,000 10,000 0.0001SD 50:33 505:03 5050:03 50500:03 505000:30 0.001 SD 5:06 50:33 505:03 5050:03 50500:30 0.01SD 0:33 5:06 50:33 505:03 5050:30 0.1 SD 0:06 0:33 5:06 50:33 505:30 SD 0:03 0:06 0:33 5:06 51:00 10SD 0:03 0:03 0:06 0:36 5:33 1005 / 2 0:03 0:03 0:03 0:09 1:03 1 000572 0:03 0:03 0:03 0:06 0:36 10 0005£> 0:03 0:03 0:03 0:06 0:33

[0131] In Table 1, we observe that when the CSOI concentration is equal to the decision threshold SD, the value of the threshold for the number of Ssor sequences must not exceed 100 if we wish to reach the threshold for the number of Ssol sequences in 30 minutes of sequencing. Indeed, the higher the threshold for the number of Ssoi sequences, the longer the analysis time increases. Logically, the higher the threshold for the number of Ssoi sequences, the longer the time to reach it is also higher. A value Ssoi = 1 is reached quickly, even for low concentrations of the species of interest. However, adopting too low a value of Ssoj increases the risk of false positives and harms the quantification accuracy.

[0132] Furthermore, a value that is too low limits the possibility of detecting other species of interest potentially present at lower concentrations. For example, the use of a sequence number threshold equal to 100 allows the detection from a single read of another species of interest SOI' present at a concentration 100 times lower. Thus, a high sequence number threshold Sso / can be particularly interesting if a majority SOI is present at a concentration greater than 100 times the SD; this allows the detection of other minority SOIs, noted SOI', at a concentration up to 100 times lower than SOI but greater than the decision threshold SD.

[0133] Based on the results presented in Table 1, it is concluded that the threshold for the number of Sso! sequences is preferably between 5 and 500 or between 50 and 500. When a short analysis time is preferred, a relatively low threshold for the number of Sso[ sequences is chosen, close to 50, or even lower, knowing that a threshold for the number of sequences that is too low is conducive to false positives. When measurement dynamics and the possibility of simultaneously detecting different species of interest, present in different concentrations, are preferred, a higher threshold for the number of Ssoi sequences, close to or greater than 500, is chosen, to the detriment of the analysis time.

[0134] The influence of the control threshold Sspc on the implementation of the method was studied. In the absence of the species of interest, significant quantities of DNA may be present in the sample. In a clinical sample, the DNA may have various origins, for example DNA from the patient's cells and from the commensal flora. The DNA may also result from an atypical infection: the concentration of genomic DNA from a microorganism other than the SOI may therefore reach quantities well above the decision threshold SD. Table 2 gives, for different concentrations of DNA CADNj expressed relative to the decision threshold SD, the sequencing time allowing to obtain a number of sequences corresponding to the control species higher than the control threshold Sspc. It was assumed that the concentration of species of interest CSO1 corresponds to the decision threshold SD.The objective is to determine the effect of varying the Sspc control threshold on the sequencing time of a sample in which the SOI is not detected.

[0135] [Tables2] DNA 0:12 0.1 SD 0:03 0:03 0:03 0:03 0:12 SD 0:03 0:03 0:03 0:06 0:21 10SD 0:03 0:03 0:06 0:21 1:45 1QQSD 0:03 0:09 0:39 3:12 15:48 1000SD 0:18 1:18 6:18 31:18 156:27 W000SD 2:33 12:33 62:33 312:33 1562:42

[0136] It is observed that a Sspc control threshold between 5 and 3125 is reached in less than 30 minutes of sequencing when the DNA concentration is equivalent to 5'^ (CADn ~ 579). It is therefore possible to conclude for these samples in the absence of SOI in less than 30 minutes of sequencing.

[0137] In cases where the DNA concentration exceeds SD, the sequencing time increases significantly, it is therefore necessary to choose a control threshold that is not too high to be able to return a negative result for the detection of the species of interest SOI. For example, in the case of a clinical sample having an atypical infection with a concentration of the pathogen equivalent to 100051D, a control threshold 5^-=125 does not allow a negative result to be returned for the detection of the species of interest SOI before at least 6 hours of sequencing. However, it is recommended not to choose a threshold for the number of Ssoi sequences that is too low to allow the sequencer to produce enough reads to allow the detection of SOI and to identify another species, different from the species of interest, but whose presence at a concentration higher than SD may be interesting to report. This situation can occur in the context of an atypical infection.

[0138] Determination of the added concentration of Cspc control species

[0139] In the preceding paragraphs, the control species SPC was present at a concentration Cspc equal to the decision threshold SD. The inventor evaluated the effect of varying the added concentration Cspc of the control species.

[0140] Table 3 indicates for different concentrations of SPC added in the sample (Cspc= 0.01SD; Cspc= SD and Cspc= 100579) the time (HH:MM) necessary for the result rendering as well as the detection result for different Cso / SOI concentrations between 0.0001SD and 10,000 SD.

[0141] In the “Results” column:

[0142] - The presence of concentrations expressed relative to the SD decision threshold in the “Results” columns indicates that the detection of SOI ( > 0) and SPC (Ni q) reads allowed the calculation of this concentration. SOI SPC This corresponds to step 62 or step 64 with > Q - The mention "NEG" indicates that no reads have been associated with the SOI = 0 ) but that the number of reads associated with the SPC exceeds the control threshold set at 25 (^y', . > 95). This corresponds to step 64 with jy', _ q. SPC SOI - The mention ">SPC" indicates that no read has been associated with the SPC (N\PC = 0) but that the number of reads associated with the SOI (n'SO{ > 1()()) exceeds the threshold of the number of Ssoi sequences set at 100. Under these conditions it is not possible to calculate the concentration CSOj[ of the species of interest, but the latter can be considered as being higher than the concentration of the control species Cspc. This corresponds to step 63.

[0143] First of all, a low Cspc concentration was taken into account, equal to 0.0151). When C soi > 0.1 SD, we find substantially the same durations as in Table 1. Thus, the added quantity of SPC does not influence the duration to reach the threshold of the number of Ssoh sequences. However, the absence of SPC reads does not allow the calculation of the SOI concentration, and the mention ">SPC" is not sufficient to allow the comparison of this concentration with respect to SD. When the CSOI concentration < OO LS'D. it is necessary to sequence more than 12 hours to be able to return a CSOI < SD result. In this context, if the total sequencing duration is limited to 12 hours, it is not possible to return negative results for these samples.

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150]

[0151] [Tables 3] SPC 0.01SD SD 100SD Csor TTR Results TTR Results TTR Results 0.0001 SD 12:33 0.0001 SD 0:09 NEG 0:03 NEG 0.001 SD 12:33 0.001 SD 0:09 NEG 0:03 NEG 0.01SD 12:33 0.01SD 0:09 0.008SD 0:03 NEG 0.1SD 5:03 >SPC 0:09 0.09SD 0:03 0.05SD SD 0:33 >SPC 0:09 SD 0:03 SD 10SD 0:06 >SPC 0:06 10SD 0:03 10SD 100SD 0:03 >SPC 0:03 100SD 0:03 100SD 1000SD 0:03 >SPC 0:03 1000SD 0:03 1000SD 10000SD 0:03 >SPC 0:03 >SPC 0:03 10000SD A Cspc concentration equal to the decision threshold increases the detection limit compared to a Cspc concentration < 0.01 SD. However, this allows quantification of the concentration of the species of interest with greater dynamics: between 0.01*5^ and 1000*5^. A high Cspc concentration, equal to 100 times the SD decision threshold, was then considered. The high amount of SPC allows for very rapid results, as in the range of SOI concentrations evaluated, the results are returned only 3 minutes after the start of sequencing. However, an increase in the detection limit of at least 10 times is observed compared to tests carried out with a concentration Cspc=SD. This can be understood by the fact that the high presence of SPC reduces the number of reads associated with the SOI. Thus, the sensitivity is reduced. Adding a large amount of SPC also allows a quantitative result to be given over a wider range than for lower concentrations. It is therefore possible to select the different analysis parameters, such as: - the Cspc concentration of the control species added to the sample; - the threshold of the number of Ssof sequences; - the Sspc control threshold; The parameterization allows to adapt the performance of the previously described process in terms of sequencing speed, quantification dynamics and the required sensitivity. For example, in the context of a clinical diagnosis requiring rapid detection with high detection sensitivity as well as precision in the quantification of SOIs detected around the clinical decision threshold, it is advantageous to add the SPC. at a concentration close to the clinical decision threshold Cspc "SD" (for example between Ol^D and 10 SD or even between 0.01 SD and 100^-0) with a threshold for the number of Ssol sequences between 5 and 500 and a control threshold Sspc between 5 and 125. Example 1

[0152] In a first example, Bacillus subtilis was used as a control species for metagenomic sequencing of samples resulting from bronchoalveolar lavage (BAL) performed on human patients. It is known that this type of sample is likely to contain a significant amount of human DNA from the patient. 38 samples were sequenced. The maximum sequencing time was set at 12 hours. Each iteration grouped 4,000 reads.

[0153] The volume of each sample was 600 pL. The control species was added at a concentration of 1.7E4 CFU / mL, CFU meaning Colony-Forming Unit. This threshold is considered slightly above a clinical threshold separating a normal presence of bacteria from a presence of bacteria representative of an infection.

[0154] In order to eliminate the patient's DNA, the analysis protocol involves eliminating the patient's DNA during a first lysis. During the first lysis, the sample was treated with a lysing agent specifically targeting the patient's cells. Such a lysing agent is for example described in WO2014 / 114896. The released DNA was then eliminated by enzymatic action and washing. The sample was then subjected to a second lysis, mechanical and chemical, in order to extract the bacterial DNA. The mechanical lysis was carried out using agitation, lasting 20 minutes, using 1 mm diameter glass beads and 0.1 mm diameter Zr / Si microbeads. The DNA was extracted from the lysate using the easyMAG platform (Biomérieux). The resulting elution volume was 25 pL. DNA extracts were stored at -20°C.

[0155] The preparation of the sequencing libraries and the barcoding of the DNA fragments were carried out using the "Rapid PCR Barcoding kit" (Oxford Nanopore Technologies). 4 samples were processed in parallel. The sequencing was carried out with the MinlON sequencer (Oxford Nanopore Technologies), connected to a MinIT computer (Oxford Nanopore Technologies). The sequencing data were processed by the WIMP (What is in my Pot?) software of the EPI2ME suite (Oxford Nanopore Technologies). This software allows a separation of the sequences of each sample using barcodes (barcoding by molecular marker, each sample being assigned a label allowing it to be identified), an identification of the bacterial or viral or fungal species corresponding to each sequence read, as well as a count of the number of reads associated with each of the identified species.

[0156] Steps 40 through 70 were repeated for series of 4,000 reads. The detection thresholds respectively associated with the strain of interest and the control strain were 100 reads and 25 reads. Twenty different species of interest were taken into account: Escherichia coli, Klebsiella oxytoca, Klebsiella pneumoniae, Klebsiella aerogenes, Enterobacter cloacae, Serratia marcescens, Proteus mirabilis, Proteus vulgaris, Hafnia alveitrobacter, Citrobacter, Ciucteric koseri, Morganella morganii, Providencia stuartii, Pseudomonas aeruginosa, Stenotrophomonas maltophilia, Acinetobacter baumannii, Légionella pneumophila, Haemophilus influenzae, Staphylococcus aureus et Strep-tococcus pneumoniae.

[0157] When the concentrations of each species of interest were quantified, the concentrations were compared to a metagenomic threshold of 5.53 GEq / mL. This is a decision threshold, allowing one to conclude on the pathological presence of a species of interest in a sample resulting from BAL.

[0158] All 38 samples were analyzed. When at least one SOI species of interest was detected, the median sequencing time was 28.5 minutes per sample. This corresponds to the sequencing time required to achieve the condition N\OJ > 100 for at least one of the species of interest.

[0159] In the absence of detection of at least one of the species of interest, < 100 ct N' > 25)' 'the median sequencing time was 2 hours and 13 minutes. The condition SPC > Sspc helps ensure that sequencing has been completed properly performed. This ensures that a negative sample is indeed a true negative.

[0160] The results of sequencing using the method described above were compared with sequencing of the same type, carried out over a fixed period of 12 hours. Table 4 summarizes the main results obtained. The performance of each method was compared by analyzing each sample using the reference method: microbial culture.

[0161] [Tables4] 12-hour sequencing Invention True positives 19 19 False positives 48 38 False negatives 4 4 True negatives 689 699 Sensitivity 82.6% 82.6% Specificity 93.5% 94.8% Positive predictive value 28.4% 33.3% Negative predictive value 99.4% 99.4%

[0162] Table 4.

[0163] The sensitivity, specificity, positive predictive value (usually designated PPV: Positive Predictive Value), negative predictive value (usually designated NPV: Negative Predictive Value) result from the number of quantities of true positives, false positives, true negatives, false negatives.

[0164] It is observed that the method has equivalent performances, or even better than those obtained with 12-hour sequencing. It allowed the detection of all infections with a median time saving greater than 11 hours and 30 minutes. Furthermore, in the absence of infection by all the species of interest, the detection of the SPC makes it possible to confirm the correct functioning of the method in a median duration of 2 hours and 13 minutes and to produce a negative detection result.

[0165] Four false negatives were observed, however these detection failures are not attributable to the method described. 2 of the four false negatives were not detected by PCR, so it appears that these are false positives from culture or an error in species identification by the reference method. The other two false negatives correspond to the species S. marcescens which produced a number of reads lower than 100 and which were quantified at concentrations lower than the clinical decision threshold. These two discrepancies between the method and the reference method were confirmed by PCR. Increasing the sequencing time to 12 hours does not improve the detection of these two S. marcescens infections and suggests an error induced by the WIMP software used for sequence assignment.

[0166] It is observed that the method allows an improvement in the performance of the test, by lowering the number of false positives, which improves the specificity and the positive predictive value. This is due to the timely stopping of the sequencing when sufficient relevant information is available, i.e. at the moment when °U when rfspc ' . As soon as one of these thresholds is crossed, the iterations stop and a 'SPC

[0167] quantification of the concentration (or minimum concentration) of the species of interest can be carried out. It is observed that it is not useful, or even desirable, to continue the iterations up to the maximum number of iterations initially set, which corresponded to a total duration of 12 hours. Indeed, obtaining too many sequences of the species of interest can induce false positives. Example 2 In a second example, the same protocol as described in connection with the first embodiment was used. 11 BAL samples, prepared as previously described, were analyzed. The 11 samples had been previously analyzed by bacterial culture and had been found to be positive for at least one species of interest, with a concentration greater than 1E4 CFU / mL. The 11 samples were analyzed to identify at least one of the 20 species of interest previously listed. Iterative steps 40 to 80 were implemented. Of the 11 samples, 6 were considered positive for at least one species of interest in less than 30 minutes of se- quencing.

[0168] However, a quantity of 100 sequences for each detected species of interest is too low to achieve a fine characterization of their genome. Iterative steps 100 to 120 were implemented on the samples, in order to identify the presence of possible antibiotic resistance markers in the genome. Iterative steps 100 to 120 were carried out until a total sequencing time of 12 hours was obtained.

[0169] On table 5:

[0170] - the first column (ref) corresponds to the reference of each sample; - the second column (SOI) corresponds to each species of interest detected; - the third column (method) corresponds to the detection method: cultivation or implementation of the invention; - the fourth column (CSoi) corresponds to the measured or estimated concentration of each bacterial species detected, - the fifth column (depth) corresponds to the sequencing depth, as previously described. - the sixth column (TTR ID) corresponds to the duration, according to the format “hour:minute” of steps 40 to 80, for confirmation of the presence of the species of interest in the sample. It is observed that the duration is between 3 minutes (sample Cl-026) and 8 hours 37 (sample Cl-060). 6 samples were considered positive in less than 30 minutes. - the following columns correspond to the presence of a marker of antibiotic resistance to 21 antibiotic agents. The letter R indicates resistance observed to the antibiotic by culture or predicted by the presence of a marker, while the letter S indicates sensitivity observed by culture. It is not possible to pronounce on the sensitivity to an antibiotic based on the absence of resistance markers to it. This is the reason why there is no letter S in the lines corresponding to the implementation of the invention. A grayed-out letter R indicates the presence of an antibiotic resistance marker revealed by sequencing, but not confirmed by bacterial culture (grayed-out letter S): this is therefore a false positive resulting from sequencing. A white letter R on a black background indicates the presence of an antibiotic resistance marker revealed by bacterial culture, but not detected by sequencing: this is a false negative resulting from sequencing. There were 9 false negatives.

[0171] Of the 9 false negatives (white letter R on black background), 8 result from samples for which the sequencing depth is low, i.e. less than 30. It is therefore likely that the false negative rate can be reduced by continuing sequencing until a sequencing depth threshold is reached, the latter being able to be equal to 30 in this example.

[0172] It is understood that a notable advantage of the invention lies in the possibility of obtaining, as the iterations progress, increasingly precise information regarding the composition of the sample. When it comes to determining the presence of species of interest, the invention makes it possible to optimize the analysis time, by stopping the iterations as soon as a biological species of interest is detected in a sufficiently reliable manner. The analysis time is thus shortened, without loss of sensitivity.

[0173] [Tables5]

[0174]

[0175] As shown in example 2, following detection of a species of interest, additional iterations can be carried out, so as to obtain information more precise regarding the genome of each species of interest detected.

[0176] Although described with sequencing carried out using a MinlON sequencer (Oxford Nanopore Technologies), the invention can be implemented using other types of sequencers allowing real-time reading and analysis of DNA fragments present in a sequencing library or directly in the sample.

[0177] A method has been described in which the steps following the sequencing are implemented by a processing unit connected to the sequencer, in particular a unit based on processor(s) or microprocessor(s) and computer memory storing in the form of computer-executable instructions all of steps 40 to 130 and configured to store the results of the method and the intermediate values, parameters and results. This unit is advantageously connected to a screen for displaying the results of the method according to the invention to the user, this screen possibly consisting of a screen of a mobile device (e.g. smartphone) connected to the processing unit to receive and display the results. These steps can be implemented remotely, for example on a remote server connected to the sequencers, either directly or through a laboratory computer system.These steps can also be implemented on a mobile device (e.g. smartphone, tablet) connected to the sequencer, for example a MinlON, the assembly consisting of the MinlON and the mobile device being easily transportable. These steps can also be implemented by different processing units. For example, one unit, connected to the sequencer, implements the bioinformatics steps (signal readings, translation into nucleotide bases, assemblies, quality control, etc.) and another processing unit, connected to the first, implements the rest of the process.

Claims

Claims

1. Method for detecting a biological species of interest (SOI) potentially present in an analysis sample, the biological species of interest having a known or partially known genome, the sample comprising a mixture of different biological species, the method comprising the steps: - a) addition of a control species (SPC) into the sample, the control species having a known genome, the control species being added at a known concentration (CSpc) in the sample; - b) extraction of nucleic acids from the sample; - c) sequencing a portion of the nucleic acid sequences extracted during step b) using a real-time sequencer for a predetermined duration or until a predetermined number of sequencing operations is obtained; - d) following step c), assignment of WSOI, nlSPC sequences resulting from step c) to the species of interest as well as to the control species; - e) updating the quantities of sequences (NlSO{ N^p^ res" respectively associated with the species of interest and the control species, such that the quantities of sequences associated with the control species and the species of interest are: • during a first iteration of step e), the quantities of sequences assigned to the control species and the species of interest during step d) added to initial quantities of sequences ( Nson ' • during each reiteration of step e), the quantities of sequences ( / î^, n^p^ assigned to the control species and to the species of interest during step d) added to the quantities of sequence N^pC^ associated with the control species and to the species of interest resulting from the previous iteration; - f) comparison of the quantity of sequences associated with the species of interest updated in step e) with a threshold of the number of sequences (Sso}) and / or comparison of the quantity of sequence associated with the control species) m'sc a Jour in step e) with a control threshold (^spc); - g) based on at least one comparison made during step f), detection of the presence of the biological species of interest in the sample or moving on to step h); - h) repetition of steps c) to g) until an iteration stopping criterion is reached.

2. The method of claim 1, wherein step d) of each iteration is performed for a predetermined duration or until a predetermined number of sequences have been sequenced.

3. Method according to any one of the preceding claims, in which: - during step f) of an iteration, when the quantity of sequences associated with the species of interest (NSoi) is greater than the threshold of the number of sequences (Ssol); - during step e) of the iteration, when the quantity of sequences associated with the control species (NSPc) is strictly greater than zero; step g) comprises an estimation of a concentration of the species of interest (CSoi) as a function of: - the quantity of sequences associated with the species of interest (during the last update; - the quantity of sequences associated with the control species (l°rS 'a last update; - the added concentration of the control species (CSPc)-

4. Method according to any one of the preceding claims, in which: - during step f) of an iteration, when the quantity of sequences associated with the species of interest (A^OP cst greater at the threshold of the number of sequences (Ssoi)', during step e) of the iteration, when the quantity of sequences associated with the control species (]yl) is zero;

5.

6. during step g), the concentration of the species of interest (Csoi) is considered to be higher than the concentration of the control species (Cspc)- A method according to any preceding claim, wherein: during step f) of an iteration, when the quantity of sequences associated with the species of interest (^Ol) is lower at the threshold of the number of sequences (^soi)', during step f) of the iteration, when the quantity of sequences associated with the control species (^) is greater than the control threshold (Sspcy during step g), if the quantity of sequences associated with the species of interest (N^j) is zero, the concentration of the species of interest is considered to be zero. A method according to any preceding claim, wherein: - during step f) of an iteration, when the quantity of sequences associated with the species of interest (N1^) is lower than the threshold of the number of sequences (Ssoiy - during step f) of the iteration, when the quantity of sequences associated with the control species (^) is greater than the control threshold (Sspcy during step g), if the quantity of sequences associated with the species of interest ) is non-zero, step g) includes an estimation of a concentration of the species of interest (CSoi) as a function of: - the quantity of sequences associated with the species of interest (Nsoi^ of the quantity of sequences associated with the control species ( N SP(^ ' of the added concentration of the control species (CSpc)-

7. A method according to claim 3 or claim 6, wherein the concentration of the species of interest is estimated from a ratio: N'sü! Nspc ^xC ^SELF SPC Or : L$pc and ^\o / are the genome lengths of the control species and the species of interest, respectively; N* and are respectively the quantities of sequences respectively associated with the species of interest and the control species resulting from the last update; Cspc is the concentration of the control species added to the sample.

8. Method according to any one of the preceding claims, in which following the achievement of the iteration stopping criterion, - if the quantity of sequences associated with the species of interest) is less than the threshold of the number of sequences (Ssoj); - and if the quantity of sequences associated with the control species (is greater than the control threshold (Sspc); - the method generates information according to which neither the presence nor the absence of the species of interest can be confirmed.

9. Method according to any one of the preceding claims, comprising taking into account a decision threshold (SD), the method aiming to compare a concentration of the species of interest with the decision threshold, the method being such that during step a), the concentration of the control species is between 0.01 times and 100 times the decision threshold.

10. Method according to any one of the preceding claims, wherein in step h), the iteration stopping criterion is reached when: - the quantity of sequences associated with the species of interest exceeds the threshold of the number of sequences (Ssoly - and / or the quantity of sequences associated with the control species exceeds the control threshold (Sspcy - and / or the cumulative duration of steps c) reaches a predetermined maximum duration; - and / or the number of cumulative sequencings during steps c) reaches a predetermined maximum number; - and / or the number of iterations reaches a predetermined maximum number of iterations.

11. A method according to any preceding claim wherein, following detection of a presence of the species of interest in the sample, the method comprises the steps: - i) sequencing a portion of the nucleotide sequences extracted during step b) using the real-time sequencer for a predetermined duration or until a predetermined number of sequencing operations is obtained; - j) following step i), assignment of sequences ( n\Oi ) resulting from step i) to the detected species of interest; - k) updating the quantities of sequences associated with the species of interest such that the quantities of sequences associated with the species of interest are: • during a first iteration of step k), the quantity of sequences assigned to the species of interest detected during step j) added to the quantity of sequences (resulting from the last iteration of steps c) to g); • during each reiteration of step k), the quantity of sequences assigned to the species of interest ( n'SO} ) during step j) added to the quantity of sequences resulting from the previous iteration of step k); - 1) repetition of steps i) to k) until reaching an iteration stopping criterion.

12. A method according to claim 11, wherein steps i) to k) are repeated until a predetermined number of iterations is reached.

13. Method according to claim 12, in which: - during each iteration of steps i) to k), following step k), the method comprises a determination of a sequencing depth of the biological species of interest, the sequencing depth corresponding to a ratio between the cumulative length of the sequences associated with the species and the length of the genome of the species; - steps i) to k) are repeated until a predetermined sequencing depth is reached.

14. Method according to any one of claims 11 to 13 comprising, following the stopping of the iterations of steps i) to k), a detection of a typical sequence of the genome of the species of interest, the typical sequence comprising an antibiotic resistance marker or a virulence marker of the species of interest.

15. A method according to any preceding claim, wherein the sequencer is configured to perform successive sequencing of different nucleic acid sequences, one after the other, the sequencing of each sequence being considered to be performed in real time.

16. Device (1) for metagenomic analysis of a sample, comprising: - a fluidic chamber (2), intended to receive the sample; - a sequencer (3) configured to carry out real-time sequencing of the sample; - a processing unit (4), comprising a bioinformatics module (4.1), programmed to assign sequences resulting from the sequencer to a species, and a control module (4.2), programmed to implement steps c) to h) of a method according to any one of the preceding claims.

17. A computer-readable or downloadable recording medium comprising instructions for carrying out steps c) to h) of a method according to any one of claims 1 to 15.

Images