Device and method for multiplexed detection of nucleic acid sequences

Abstract

Device and method for multiplexed detection of nucleic acid sequences A device for multiplexed detection of nucleic acid sequences comprises a thermal cycler (4) arranged to perform a series of thermal cycles with an in vitro nucleic acid amplification reagent containing at least two fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores emitting in overlapping fluorescence wavelength ranges, a light sensor (6) arranged to measure radiation emitted by said fluorophores, and an analyzer (8) arranged to determine, for each respective fluorescence probe, a value representative of a concentration,from time signatures taken from the fluorescence measured as a function of time for a given thermal cycle of a reaction mixture comprising one or more probes and at least two measurements made during at least some of the thermal cycles. Fig.1,

Description

Title of the invention: Device and method for multiplexed detection of nucleic acid sequences

[0001] The invention relates to the field of detection of nucleic acid sequences by PCR reaction, or any other method of amplification of nucleic acids in vitro, and in particular the multiplexed detection of nucleic acid sequences by polymerase chain reaction (PCR).

[0002] In many situations in the field of molecular biology, it is necessary to identify with a high level of confidence the presence of one or more DNA or RNA sequences of interest in a sample, and to measure their respective relative concentration, these sequences appearing in a set of sequences of interest.

[0003] In these situations, it is often interesting to be able to test the presence of several sequences simultaneously in the same sample in a single test. The causes of this need may be urgency, the small size of the sample, the low concentration of the sequences of interest in this sample, or the cost of the test (reagents, consumables, occupancy rate of the machine, etc.).

[0004] When this or these sequences are in quantities too small to be characterized, it is conventional to implement an in vitro exponential amplification of these sequences, for example by a polymerase chain reaction (PCR). This method aims to obtain a sufficient quantity of the desired sequences using a mixture of oligonucleotide primer pairs located in these sequences of interest. Several variants of PCR have been developed (for example LATE-PCR, ASPCR). Other methods for amplifying nucleic acids in vitro are also known, such as the NASBA method (nucleic acid sequence-based amplification), the TMA method (transcription mediated amplification), LAMP (loop-mediated isothermal amplification), SDA (strand displacement amplification) and rolling circle amplification.

[0005] Various methods are already known for identifying the presence of one or more nucleic sequences of interest, after having amplified them where appropriate. It is possible to carry out sequencing without a priori, or targeted, of the nucleic acids present in the sample. Such methods are generally long and expensive and remain limited to certain applications. The most widely used methods consist of using oligonucleotide probes which can hybridize to a characteristic portion of their sequence of interest. These probes can be implemented according to several families of methods:

[0006] - DNA chip: this involves fixing the various probes on a solid support, by example a 2-dimensional matrix, and locate them by their coordinates in this matrix (principle also called "DNA microarray" in English). This inhomogeneous phase detection has a high manufacturing cost and slow hybridization kinetics which harms performance in terms of sensitivity,

[0007] - liquid chip: this involves ordering the probes on a liquid chip, see for example Luminex's xMap process published at https: / / www.luminexcorp.com / xmap-technology / . This process is complex to use because it requires a flow cytofluorimeter, and the analysis it implements is not instantaneous because the beads are analyzed sequentially,

[0008] - homogeneous phase format: the probes can consist of an oligonucleotide single nucleotide or the combination of two oligonucleotides, free in solution. Probes are conventionally labeled using a fluorophore and a quencher. When a probe is mixed with a sample that does not contain its target sequence, and the temperature is within the temperature range compatible with its hybridization, the fluorescence of this probe is reduced or eliminated by the spatial proximity of the quencher and the fluorophore. On the other hand, when it interacts with its target sequence, the fluorophore and the quencher are kept at a distance and the probe becomes fluorescent.

[0009] The homogeneous phase format can be used in several variants, the most used of which are detailed below:

[0010] - the molecular beacon: in this variant, the probe consists of a single oligonucleotide cleotide, designed so that the fluorophore and quencher are in close proximity when the probe is not hybridized and is at low temperature. This proximity most often results from the presence of complementary sequences of a few bases at the ends of the specific sequence of the sequence of interest. In solution, these two complementary ends hybridize to form a double helix, which has the effect of bringing the fluorophore and the quencher closer and keeping them at a short distance. When the target sequences are present in the solution and the temperature is below or near the melting temperature of the duplex formed between the probe and its target, the probes that hybridize to their target sequences become fluorescent, and

[0011] - the 5' nuclease activity test: in this variant, when the probe is hybridized to its target sequence, hybridization of a primer 5' to this probe triggers the polymerization of a new strand of DNA. When the polymerase encounters the 5' end of the probe, it cleaves it using its 5' to 3' exonuclease activity, which has the effect of irreversibly separating the fluorophore from the quencher. The fluorophore can then express its fluorescence more efficiently since the radiation is not more absorbed by the quencher. The principle of the method exploiting the 5' to 3' exonuclease activity was described in US patent 5,210,015 and in a publication by Holland PM et al. "Detection of specific polymerase chain reaction product by utilizing the 5'—3' exonuclease activity of Thermus aquaticus DNA polymerase", Proceedings of the National Academy of Sciences Aug 1991, 88 (16) 7276-7280; DOI: 10.1073 / pnas.88.16.7276 PNAS 1991 in 1991. In this method, the probe was labeled at the 5' end using a radioactive label. The method was then improved by replacing this label with a fluorophore with the addition of a 3' quencher which allows the probe to undergo a change in fluorescence depending on whether it is free in solution, or hybridized, or digested (US patent 5723591, probes usable with the Taq-Man™ method).

[0012] These formats can themselves be implemented according to variants such as methods using probes (which can also act as primers in some of these methods) of the Scorpion (registered trademark), Amplifluor (registered trademark), MGB Eclipse (registered trademark), Light Upon extension (LUX, registered trademark), Quenching of Unincorporated Amplification Signal Reporters (QUASR) or QZyme (registered trademark) type for example.

[0013] The homogeneous phase format has many advantages, including: excellent kinetics of interactions of the probes with their respective target sequences, low cost, and excellent performance in terms of sensitivity.

[0014] However, performing multiple measurements simultaneously is limited by the ability to separate the signals from different probes mixed in the same test, which is a major drawback. Indeed, probes targeting distinct sequences must have emission wavelengths that can be separated from each other in order to be able to use them in the same measurement.

[0015] Indeed, whatever the variant chosen (molecular beacon or 5' nuclease activity assay or other), it is not possible to use more than one probe per fluorescence channel. Commercially available real-time thermocyclers measure a fluorescence level for each channel at a single time in each cycle of the amplification reaction. Consequently, if two probes emitting in the same fluorescence channel are mixed, it is not possible to distinguish their respective contribution to the production of the fluorescence signal.

[0016] Thus, a real-time thermocycler capable of measuring 4 fluorescence channels can only detect and discriminate the fluorescence emitted by a maximum of 4 probes and the amplification reactions carried out with this device cannot detect more than 4 target sequences per reaction.

[0017] Several methods have been proposed to circumvent this limitation:

[0018] - discrimination by decoupling excitation and emission: in this process, Probes emitting in the same fluorescence channel can be mixed provided they can be excited separately. Some fluorophores can indeed be excited at wavelengths far from each other, but emit in the same wavelength band. This is particularly the case for fluorophores that emit with a large Stokes shift. Thus, Cepheid recently improved its GeneXpert (registered trademark) instrument by expanding the number of fluorophores detected from 6 to 10 (see Chakravorty et al., “Detection of Isoniazid-, Fluoroquinolone-, Amikacin-, and Kanamycin-Resistant Tuberculosis in an Automated, Multiplexed 10-Color Assay Suitable for Point-of-Care Use,” 2017, Journal of Clinical Microbiology, 55, 183, published at https: / / doi.org / 10.1128 / JCM.01771-16).In the orange emission band, for example, 3 fluorophores can thus be discriminated (CF9, excitable in blue, CF8, excitable in green, CF4, excitable in yellow). This method requires an optical module to analyze the spectral composition of the emitted signal: excitation in green produces signals emitted by the fluorophores CF3, C8 and CF10 in yellow, orange and infrared respectively. It also has the disadvantage of methods using several fluorescence bands, namely the optical contamination of a band, or channel, in an adjacent band. These optical contaminations (also called “crosstalks”) can disrupt the analysis of the results in the adjacent bands and be the source of erroneous results, such as false positives for example.

[0019] - discrimination by the melting curve method: it is known to distinguish . probes sharing the same fluorescence emission range by distinguishing the probes by their melting temperature. To do this, a melting curve is produced after initiating the formation of the duplex between the sequence of interest and the oligonucleotide probe(s), by gradually and monotonically increasing the temperature T of the solution, while recording the fluorescence signal F. The maximum of the -dF / dT curve indicates the melting temperature Tf of the duplex. This melting temperature makes it possible to ensure that the duplex formed is a homoduplex, or even a heteroduplex if this temperature is lower than the known temperature of the homoduplex, or even to identify, in this case, the mutation(s) at the origin of the heteroduplex when the possible mutant sequences are listed in a restrictive list and their respective melting temperatures with the probe are previously known in a one-to-one manner.

[0020] With respect to the melting curve method, some of the methods for detecting sequences of interest described above are incompatible with the conditions necessary for measuring the melting temperature.

[0021] This is particularly the case for the following processes:

[0022] - amplification with a polymerase having a 5'->3' exonuclease activity: this is the most common case, as with the 5' nuclease activity assay method, non-mutated polymerases generally having such activity. This results in hydrolysis of the 5' end of the probes when it is not blocked, and the hydrolyzed probe molecules are no longer usable for producing the melting curve.

[0023] - asymmetric amplification: in the general case of exponential amplifications dependent on primers like PCR, the amplification is said to be symmetrical and these primers are in stoichiometric concentration, which means that the amplification generates as many sense DNA strands as antisense. By construction, one of these strands shares the same sequence, or a very close sequence, to that of the probe, so that this strand can compete with the probe to hybridize to the target sequence, and suppress the signal emitted by this probe. To overcome this difficulty, an asymmetric amplification can be carried out by modifying the ratio of the concentrations of the primers in order to favor the formation of the DNA strand complementary to the probe, but the amplification then quickly ceases to be exponential when the primer in small quantity has been consumed to become only linear, which lengthens the duration of the test and / or reduces its sensitivity,

[0024] - multiplex amplification with a universal primer: here we use primers containing a unique conserved region in their 5' region. This makes it possible to advantageously and efficiently reduce the occurrence of primer dimers (Brownie et al "The elimination of primer-dimer accumulation in PCR", Nucleic acids Res. 1997 Nucleic Acids Res. 1997 Aug 15;25(16):3235-41. doi: 10.1093 / nar / 25.16.3235, published at https: / / doi.org / 10.1093 / nar / 25.16.3235.) and to exponentially amplify a large number of sequences of interest with a single primer pair. On the other hand, it is obviously impossible to carry out asymmetric amplification, since the primer used amplifies each of the sense and antisense strands of the sequence of interest.

[0025] In the case where the probe is not hydrolyzed, its hybridization can be suppressed or considerably reduced by the presence of the complementary strand formed during the symmetrical amplification reaction. This reduction in the signal can result from at least two mechanisms:

[0026] - competition during hybridization with this complementary strand, which, even in concentration lower than that of the probe, can hybridize first and preferentially,

[0027] - displacement of the probe by the complementary strand, which can hybridize in homologous regions in 5' or 3' of the sequence of interest, even after hybridization of the latter. This complementary strand can then displace the probe by branch migration ("toehold-mediated strand displacement" in English) according to kinetics which depend in particular on the length of these homologous regions (see the article by Srinivas et al. "On the biophysics and kinetics of toehold-mediated DNA strand displacement", Nucleic acids research 2013, published at https: / / doi.org / 10.1093 / nar / gkt801),

[0028] - discrimination by the TOCE and MuDT methods of the Seegene company: the The above method according to the melting curve is carried out at the end point and therefore does not allow quantification of the relative concentrations of the sequences of interest if they are simultaneously present. The company Seegene has described two methods to circumvent this limitation. The first, called TOCE ("Tagging Oligonucleotide Cleavage & Extension", registered trademark) allows detection of up to 5 probes per fluorescence emission channel using fluorescent "catcher" probes distinguished by their melting temperature. The main drawback is the complexity of implementing this method, which requires adding several oligonucleotides in addition to those used to amplify the sequences of interest. These oligonucleotides are likely to generate artifactual products that consume reagents to the detriment of the production of the desired PCR products.The second, called MuDT ("Multi Ct values in a single cell", registered trademark), allows the combination of up to 3 probes in the same excitation and emission channels. The probes have distinct melting temperatures, separated by several degrees. PCR amplification is carried out with cycles whose temperature profile includes one or more steps during the temperature increase towards the melting temperature of around 95 °C. The temperature steps are chosen between the melting temperatures of the probes, and the fluorescence signal is recorded at each of these steps. By calculation, the contribution of each of the probes to the total fluorescence signal can be determined. A major drawback is that the thermal cycler used must be able to manage this type of particular cycle profile, which is incompatible with rapid PCR in which the aim is to minimize the temperature change times during the cycles.This method is also incompatible with TaqMan mode detection in which the probe is hydrolyzed.

[0029] To date, the methods that claim to be multiplex are therefore extremely unsatisfactory, either because of their complexity, or because of their performance, or because of their incompatibility with widespread homogeneous phase formats such as TaqMan probes. For the most part, one could even describe them as multi-fluorescence rather than multiplex, since it is mainly a question of making "parallel" or simultaneous measurements, but not multiplexed, i.e. whose signals are combined with each other.

[0030] The invention improves the situation. To this end, it proposes a device for multiplexed detection of nucleic acid sequences comprising a thermal cycler arranged to carry out a series of thermal cycles with an in vitro nucleic acid amplification reagent containing at least two fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluoro rophores emitting in overlapping fluorescence wavelength ranges, a light sensor arranged to measure radiation emitted by said fluorophores in said fluorescence wavelength ranges, the radiation emitted by each probe varying depending on whether it is in an unmodified state in which the quencher substantially attenuates or does not attenuate the fluorescence emission, or in a modified state in which the quencher has an opposite effect on the fluorescence emission, and an analyzer arranged to determine, for each respective fluorescence probe, a value representative of a concentration in a modified state, from time signatures derived from the measured fluorescence as a function of time for a given thermal cycle of a reaction mixture comprising one or more probes, each of which may be substantially entirely in an unmodified or modified state,and at least two measurements carried out during at least some of the thermal cycles, which values representative of a concentration in a modified state make it possible to qualify the presence of one or more nucleic acid sequences each associated with a distinct fluorescence probe so as to cause a change of state of the fluorescence probe from the unmodified state to the modified state when they interact.

[0031] This device is particularly advantageous because it makes it possible to detect the presence of a sequence of interest using a characterized fluorescent probe whose time signature has been previously characterized. Thus, the determination of this time signature makes it possible to simultaneously use probes emitting in the same fluorescence emission band, but whose time signatures are sufficiently different so that their respective contributions can nevertheless be separated. Advantageously, this time signature also makes it possible to effectively eliminate potential optical contaminations from one fluorescence band to an adjacent band. With this device, potential optical contaminations can even be used to detect the presence of a sequence of interest.

[0032] According to various embodiments, the invention may have one or more of the following characteristics:

[0033] - the thermocycler is a PCR thermocycler, and the at least two fluorescent probes rescence are selected from the group including TaqMan probes, molecular beacons, dual fluorescence transfer probes, Scorpion probes, Amplifluor probes, MGB Eclipse probes, LUX probes, QUASR probes and QZyme probes,

[0034] - the in vitro nucleic acid amplification reagent comprises three probes of fluorescence combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges vauching,

[0035] - the in vitro nucleic acid amplification reagent comprises at least two fluorescence probe groups combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges within each fluorescence probe group

[0036] - the analyzer is arranged to determine the representative value of the concentration of each fluorescence probe in an altered state by solving a matrix system based on fluorescence measurements, time signatures and the constancy of the respective concentration of each fluorescence probe,

[0037] - the analyzer is arranged to solve a matrix system by group of probes of fluorescence,

[0038] - the analyzer is arranged to use a minimization algorithm, and

[0039] - the analyzer is arranged to determine the representative value of the concentration of each fluorescence probe in an altered state by applying gradient descent or using a neural network.

[0040] The invention also relates to a method for multiplexed detection of nucleic acid sequences comprising the following operations:

[0041] a) performing a plurality of thermal cycles on a reaction mixture comprising a sample to be analyzed and an in vitro nucleic acid amplification reagent containing at least two fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores emitting in overlapping fluorescence wavelength ranges,

[0042] b) during each thermal cycle, carrying out at least two fluorescence measurements in said fluorescence wavelength ranges, the radiation emitted by each probe varying according to whether the latter is in an unmodified state in which the quencher attenuates or does not attenuate the fluorescence emission substantially, or in a modified state in which the quencher has an opposite effect on the fluorescence emission,

[0043] c) determining a value representative of the concentration of each fluorescence probe in a modified state from the measurements of operation b) and from time signatures derived from the fluorescence measured as a function of time for a given thermal cycle of a reaction mixture comprising one or more probes, each of which may be substantially entirely in an unmodified or modified state,

[0044] which values representative of a concentration in a modified state make it possible to qualify the presence of one or more nucleic acid sequences each associated with a distinct fluorescence probe so as to cause a change of state of the fluorescence probe from the unmodified state to the modified state when they interact.

[0045] According to various embodiments, the method may have one or more of the following characteristics:

[0046] - operation b) comprises carrying out PCR cycles, and the at least two probes fluorescence probes are selected from the group comprising TaqMan probes, molecular beacons, dual fluorescence transfer probes, Scorpion probes, Amplifluor probes, MGB Eclipse probes, LUX probes, QUASR probes and QZyme probes,

[0047] - operation b) comprises the use of a bare acid amplification reagent in vitro nucleic acid sequences comprising three fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges,

[0048] - operation b) comprises the use of a naked acid amplification reagent in vitro nucleic acid sequences comprising at least two groups of fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges within each group of fluorescence probes,

[0049] - operation c) comprises the resolution of a matrix system based on the measurements fluorescence, the temporal signatures and the constancy of the respective concentration of each fluorescence probe,

[0050] - operation c) comprises the resolution of a matrix system by group of probes of fluorescence, and

[0051] - operation c) comprises the application of gradient descent or the use of a neural network.

[0052] Other characteristics and advantages of the invention will appear more clearly on reading the following description, taken from examples given for illustrative and non-limiting purposes, taken from the drawings in which:

[0053] - [Fig.l] represents a generic diagram of a device according to the invention,

[0054] - [Fig.2] shows the fluorescence intensity obtained as a function of time during two PCR reactions using a TaqMan probe labeled respectively with the fluorophore ATTO 565 and the fluorophore Cy3,

[0055] - [Fig.3] represents the time signature of a set of probes, with, for each of them, the profile of the temperature cycle used, the respective components of this signature when the probe has not interacted, and has interacted, with the target sequence for which it is specific,

[0056] - [Fig.4] illustrates the implementation of the method with the device in the mode by particular of a PCR reaction carried out with two TaqMan probes emitting in the same fluorescence band, including the decomposition into an optimal sum of the time signatures as described above,

[0057] - [Fig.5] illustrates the implementation of the method with the device in the mode by particular of a PCR reaction carried out with two TaqMan probes emitting in the same fluorescence band, including the decomposition into an optimal sum of the time signatures as described above, during a multiplex amplification of two sequences of interest,

[0058] - [Fig.6] illustrates the implementation of the method with the device for eliminating the optical contamination signals from one channel to an adjacent channel, as well as in the particular mode of a PCR reaction carried out with molecular beacon and QUASR type probes.

[0059] The drawings and the description below contain, for the most part, elements of a certain character. They may therefore not only serve to better understand the present invention, but also contribute to its definition, where appropriate.

[0060] The invention provides a device capable of carrying out multiplexed detection of nucleic acid sequences, i.e. using probes comprising one or more fluorophores allowing the identification of genetic sequences by modification of the signal emitted during the interaction of said probes with said sequences (for example TaqMan, Molecular Beacon, FRET probes, etc.), said fluorophores emitting in fluorescence wavelength ranges which may possibly overlap.

[0061] The interaction of the probe with the target sequence for which it is specific can take several forms depending on the detection method used. There are in fact several known methods, such as, without this list being limiting, detection by real-time PCR using TaqMan type probes, molecular beacons, MGB Eclipse probes, dual probes using fluorescence transfer (FRET), or detection by real-time PCR using fluorescent primers also serving as Scorpion, Amplifluor, LUX, QUASR or QZyme type probes.

[0062] The chemical nature of the probe can also vary depending on the methods. It can be made into deoxyribonucleic acid (DNA), ribonucleic acid (RNA), or synthetic nucleic acid analogs such as peptidonucleic acids (Peptide Nucleic Acids, PNA), "locked" nucleic acids (LNA). The probe can include chemical groups modifying its interaction with the genetic sequence for which it is specific, such as a group binding to the minor groove of the double helix (MGB).

[0063] According to the method implemented, the probe may consist of a single oligonucleotide molecule comprising one or more fluorophores and one or more chemical groups having the effect of modulating the intensity of their fluorescence according to the state of the probe, called "quenchers" in the rest of the text. The probe may also consist of two or more molecules, some comprising one or more fluorophores, the others comprising one or more chemical groups having the effect of modulating the intensity of their fluorescence according to the possible interactions defining the state of the probe.This is for example the case with a QUASR probe comprising two oligonucleotides, or with a dual FRET-type fluorescence transfer probe consisting of two oligonucleotides, one carrying at its 3' end a fluorophore capable of de-exciting itself by transferring its energy to a quencher carried at the 5' end of a second oligonucleotide, this quencher itself being able to de-excite itself by emitting light radiation in a wavelength band different from that of the fluorophore.

[0064] Depending on the detection method implemented, the interaction of the probe may consist of a simple hybridization with the genetic sequence for which it is specific, or a hybridization followed by hydrolysis mediated by an enzymatic activity, or a hybridization followed by incorporation into an existing DNA strand mediated by the action of a ligase, or into a synthesized DNA strand mediated by the action of a polymerase.

[0065] In the remainder of the text, the term "interact" will be used generically to describe the action of a probe which interacts with the genetic sequence for which it is specific according to one of the methods described above, and the term "modified by the interaction" will be used to describe the process by which the fluorescence characteristics of a probe are modified by its interaction with the genetic sequence for which it is specific according to one of the methods described above.

[0066] The multiplexed detection according to the invention is made possible by determining fluorescence time signatures for each type of fluorescent probe used, by acquiring two or more fluorescence signals during at least two thermal cycles applied during the amplification reaction, and by implementing an algorithm for separating the signals emitted by these probes using these signatures.

[0067] Thus, in a particular embodiment of the invention, the device applies one or more thermal cycles to these probes by exciting them optically and by continuously or sampledly recording the fluorescence signals emitted in one or more wavelength bands, which makes it possible to carry out the multiplex measurement. When the amplification method comprises a series of thermal cycles as in the case of PCR, the fluorescence signals are recorded during at least two of these cycles, and preferably, during each cycle. When the amplification method is isothermal, at least two thermal cycles are applied to the sample, one at the beginning, the other at the end of the reaction, during which the emitted fluorescence signals are recorded.

[0068] More specifically, and as shown in [Fig.l], the device 2 according to the invention comprises a thermocycler 4 capable of applying two or more temperature cycles to a reaction mixture of the amplification containing at least two fluorescent probes, a light sensor 6 having the capacity to carry out at least two fluorescence intensity measurements per cycle in at least one wavelength band, and an analyzer 8 arranged to carry out the measurement of the relative concentration of the different states in which the different probes are found during the reaction, and to deduce therefrom the presence of one or more nucleotide sequences of interest targeted by said probes. In the example described here, the light sensor 6 comprises both the excitation source of the fluorescent probes and the corresponding photodetector. Alternatively, the light sensor 6 could be separated into source and detector.In the example described here, the analyzer 8 is a suitable computer program or code executed on one or more processors. By processors, it is meant any processor suitable for calculating the projection of textures onto planes and processing linked to voxels. Such a processor can be implemented in any known manner, in the form of a microprocessor for a personal computer, a dedicated chip of the FPGA or SoC type ("System on chip" in English), a computing resource on a grid or in a cloud, a microcontroller, or any other form suitable for providing the computing power necessary for the implementation described below. One or more of these elements can also be implemented in the form of specialized electronic circuits such as an ASIC. A combination of processor and electronic circuits can also be envisaged.

[0069] The thermal cycler 4 is capable of applying variable temperatures within a given range to a sample mixed with an in vitro nucleic acid amplification reagent, this reagent containing at least two fluorescent probes having distinct time signatures. The thermal cycler 4 can repeatedly apply a given temperature profile to this mixture. The analyzer 8 receives or determines the temperature values applied to the sample. In the example described here, these values are sent to the analyzer 8. According to various embodiments, they can be measured in or in contact with the sample, or extrapolated as a function of time. In a preferred embodiment of the invention, the thermal cycler 4 is a fast thermal cycler (i.e. with a temperature change of the order of more than 5°C per second, and more preferably more than 15°C per second). This is made possible thanks to the analyzer 8 according to the invention.Conventionally, fast thermocyclers are not used to attempt to perform multiplex measurements. because the information they provide is not sufficiently precise.

[0070] The light sensor 6 is arranged to subject the reaction mixture to light excitation in one or more wavelength bands in the ultraviolet and / or in the visible and / or in the infrared, and to measure the intensity of the fluorescence emission resulting from this excitation, in one or more wavelength bands in the ultraviolet and / or in the visible and / or in the infrared, by carrying out one or more punctual measurements for a determined duration, or a series of punctual measurements at a given frequency (for example, for 100ms every 200ms). These fluorescence measurements as a function of time in the accessible wavelength band(s) are sent to the analyzer 8. In a preferred embodiment, the acquisition periodicity is between 10ms and 10s, and even more preferably between 100ms and 1s.Thus, unlike all known state-of-the-art, the analyzer 8 can monitor almost continuously the evolution of the fluorescence response throughout the thermal cycles.

[0071] In a particular and preferred embodiment of the method according to the invention, the PCR reaction is used to detect and discriminate the presence or absence of a genetic target in combination, the PCR reagent containing a set of probes combining a quencher and a fluorophore (such as TaqMan probes for example), which fluorescent probes use distinct fluorophores which can emit in a common band or range of wavelengths.

[0072] Thanks to the prior knowledge of the time signature of each of the probes, the analyzer 8 is capable of measuring unambiguously, for each probe present in the reagent, the modification of the fluorescence signal resulting from the interaction between the probe and the amplification products (or amplicons) derived from the genetic sequence for which it is specific, independently of each other, and thus of reaching a multiplexing level of 2n, or even 3n or 4n, with a device according to the invention capable of measuring the fluorescence signal in n wavelength bands simultaneously.

[0073] This method makes it possible to recognize that a sequence of interest is present in a sample, not by simply observing an increase in fluorescence at low temperature (60°C for example) over time as is conventionally done during an analysis by amplification of nucleic acids in vitro in real time, but by observing the change in the fluorescence profile measured by the light sensor 6 during a thermal cycle. [Fig. 2] shows two fluorescence curves recorded with the device 2, in the same fluorescence channel, but during PCR reactions, in which the probes used are functionalized with two different fluorophores (ATTO 565 and Cy3). These curves show that the fluorescence profiles as a function of time during a cycle vary, depending the nature of the probe, and depending on whether the cycle is at the beginning or the end of the amplification.

[0074] In a particular embodiment of the method according to the invention, the thermal cycle is a PCR cycle and comprises increasing the temperature from a low temperature to a high temperature, the low temperature being between 55 and 70°C, preferably between 58 and 65°C and the high temperature being between 80 and 100°C, preferably between 90 and 98°C, followed by decreasing the temperature from the high temperature to the low temperature.

[0075] In the present description, thermal cycles are described generically as comprising increasing the temperature from 60 to 95°C, followed by decreasing the temperature to 60°C. These values are particularly suitable for carrying out a rapid PCR reaction. It goes without saying that the values of these ranges (60°C, 95°C and 60°C) could be modified to adapt to particular thermal cycles used during PCR amplification with for example the use of 3 temperatures instead of 2, or even thermal cycles added during isothermal amplification according to a temperature profile compatible with the thermal stability of the reagents used.

[0076] In a particular embodiment of the method according to the invention, the signal modification results from the hydrolysis of the probe(s) in the case of TaqMan probes cleaved by a polymerase having exonuclease activity.

[0077] In addition, the device 2 can be used to determine the time signature of each of the probes present in the reaction mixture.

[0078] General principle of the invention

[0079] The general principle implemented by the invention is based on the detection of the modification of the fluorescence signal emitted by a probe depending on whether or not it has interacted with the nucleotide sequence of interest for which it is specific. The Applicant's work has made it possible to identify 1) that this fluorescence signal generally depends on the temperature through several more or less well-known physical phenomena which occur concomitantly, the main ones of which will be cited below; 2) that the variation of the fluorescence signal of this probe with the temperature can be measured by applying a temperature cycle to the sample containing the probe(s); 3) that the fluorescence curve obtained itself varies depending on whether or not the probe has interacted with the nucleotide sequence for which it is specific;3) that this variation in the variation of the signal during a cycle is specific to the probe and constitutes a signature which can be used to detect, at the time of application of a thermal cycle, the fraction of this probe which is found to have interacted with the target sequence for which it is specific; 4) that the determination of this fraction at at least two chosen times of the reaction; amplification can make it possible to deduce the presence or absence, or the initial concentration through the determination of a threshold cycle, in the amplification reaction mixture, of the nucleotide sequence of interest for which the probe is specific.

[0080] Recalling that in most cases, the fluorescence level of a probe interacting with a specific nucleotide sequence varies thanks to the modification of the distance between a fluorophore and a quencher (case of TaqMan, Molecular Beacon, FRET, MGB Eclipse probes, and primers acting as Scorpion, Amplifluor, LUX, QUASR, etc. probes), this variation in fluorescence can depend on the temperature in several ways.

[0081] First, the variation of the fluorescence quantum yield of the fluorophore used, which decreases as the temperature increases. For example, the yield of rhodamine B decreases monotonically with temperature from 0.8 at 10°C to 0.3 at 60°C (see the article by Kubin et al. “Fluorescence quantum yields of some rhodamine dyes”, Journal of Luminescence 1983, doi.org / 10.1016 / 0022-2313(82)90045-X). In this specific case, this decrease is mainly due to the increase in dynamic quenching with temperature (see the technical note by Arnaoutakis 2016, “Quenching of fluorescence with temperature”, Technical note from EDINBURGH INSTRUMENTS published at https: / / www.edinst.com / wp-content / uploads / 2018 / 10 / TN_27-Quenching-of-Fluorescen ce-with-Temperature.pdf). This variation in quantum yield with temperature itself varies with the nature of the fluorophore.Thus, the variation of the fluorescence quantum yield with temperature of two fluorophores emitting in the same wavelength range can differ from one fluorophore to another.

[0082] Then, the change in the conformation of the probe, which modifies the distance between the fluorophore and the quencher, or the location of the fluorophore or the quencher (for example, the fluorophore of a probe may be close to the quencher of another probe or to the quenchers of several probes) has a consequence on the fluorescence signal emitted by the fluorophore. More generally, the mixture of the amplification reaction containing the fluorescent probes and the other reagents is regulated by intramolecular and intermolecular interactions between the different chemical species. The oligonucleotides in solution can adopt different conformations depending on the temperature and the stage of the amplification reaction (the presence in increasing concentration of sequences complementary to the probes generated during the amplification reaction).As a first approximation, for an intact probe, different situations are possible at each instant of the amplification reaction: . 1) the probe interacts with a complementary sequence generated during the amplification reaction, 2) the probe does not interact with any complementary sequence, or 3) the probe is bound to complex structures, with one or more oligonucleotides in solution linked together.

[0083] The probe can therefore adopt different states, for example different conformations, depending on whether it is in one or other of these situations.

[0084] The efficiency of the quencher in quenching the fluorescence emitted by the fluorophore is a third factor influencing the variation of the probe fluorescence intensity with temperature.

[0085] Finally, the state of the probe can also be an important factor: In the TaqMan process, when the probe has hybridized to its target sequence and has been hydrolyzed in the revelation mode by the 5' to 3' exonuclease activity of the polymerase, the fluorophore is irreversibly cleaved from the probe and its fluorescence is no longer impacted by the proximity of the quencher. The interaction of the probe with the nucleotide sequence for which it is specific thus results in a change of state in which the fluorophore of the probe is no longer molecularly bound to the quencher.

[0086] The Applicant has discovered experimentally that the laws of variation of each of these phenomena with temperature can themselves vary significantly depending on the structure of the probe and in particular the nature of the fluorophore, the sequence of the probe and the nature of the quencher, as well as the proximity to another probe. Depending on the parameters, the combination of these different laws can produce an overall law that varies greatly from one probe to another. In particular, the effect of temperature on the quantum yield of the fluorophore and that of temperature on the conformation of the probe have contributions of opposite signs and different time constants.

[0087] The Applicant's work has enabled it to identify that, for a given temperature profile, this curve constitutes the first component of a time signature.

[0088] When a nucleotide sequence complementary to one of the probes is present in a sample at a certain concentration and one of the molecules carrying this sequence interacts with a molecule of this probe specific to this sequence, the law of variation of the intensity of the fluorescence as a function of the temperature of this probe can be modified, whether because the probe adopts another conformation due to its state of hybridization (as is the case for Molecular Beacon probes), because a component of the probe is close to another component of the probe having a disturbing effect (as is the case for dual FRET type probes or Scorpion primers) or because a polymerase having a 5' to 3' exonuclease activity is present in the sample, and the fluorophore of the probes which have hybridized with their complementary sequence is cleaved by this exonuclease (as is the case for TaqMan probes), the law of variation of fluorescence as a function of the temperature of this probe is modified.

[0089] The Applicant's work has enabled it to identify that, for a given temperature profile, this variation constitutes a second component of the time signature which replaces the first component of the signature.

[0090] By analyzing and comparing the fluorescence signals recorded during a given temperature cycle before and after the interactions resulting from the presence of the amplification product, it therefore becomes possible to calculate, for each probe and at the instant when the thermal cycle is applied, its fraction which has been modified by this interaction, and to thus deduce the quantity of the amplification product resulting from this specific sequence present in the sample thanks to these two time signature components.

[0091] During the first cycles or the first moments of an amplification reaction in which a certain quantity of a sequence of interest is found, the molecules of the probe for which it is specific are not significantly modified by the presence of an amplification product resulting from this sequence of interest (except in the particular case of a high initial concentration of this sequence of interest), and the recorded fluorescence signal comes solely from the component of the temporal signature of the probe not modified by the interaction with the sequence for which it is specific.After a number of cycles, an increasing fraction of the probe is modified as a result of interaction with the amplification product from the sequence of interest, and the recorded signal is a linear combination of the contributions of the respective time signature components of the unmodified and modified probe by interaction with the amplification products from the sequence of interest.

[0092] If several fluorescent probes emitting in the same fluorescence band are mixed and the fluorescence intensity emitted by this mixture of probes is continuously recorded, with a sufficient sampling frequency, during one or more cycles (before and after modification of these probes by the presence of the amplified targeted sequence(s), it is possible, thanks to their respective time signatures, to separate at each cycle the contributions to the signal of each of the probes by determining for each probe the fraction which is in unmodified form and the fraction which is in modified form by the interaction with the nucleotide sequence for which it is specific.

[0093] When probes emit predominantly in fluorescence bands that are not common but adjacent, as is the case with many organic fluorophores, the method makes it possible in particular to eliminate optical contaminations from a band to band (“crosstalks” in English) or any other variation in fluorescence not resulting from the modification of the probe by the interaction with the sequence for which it is specific. In the state of the art, the elimination of these optical contaminations is conventionally carried out by the application of a linear correction using a square matrix whose non-diagonal terms depend on the identity of the fluorophores used in a multiplex kit. For example, with a thermal cycler having 4 fluorescence detection channels, the user enters the identity of the fluorophores used in his kit. This indication makes it possible to select an optical contamination correction matrix adapted to this combination of fluorophores.

[0094] In the method according to the invention, such a correction with a matrix is no longer possible since several fluorophores emitting predominantly in a channel A can contaminate, at different rates, the signal in an adjacent channel B. Thus, the contamination rate generated by all of these fluorophores in channel B will depend on the amplification or not of each of the sequences of interest detected in channel A. In the method according to the invention, the temporal signatures of this probe in the adjacent channel(s) are determined or known for a probe emitting predominantly in a channel A. And these signatures are used to precisely determine their contribution to the signal in this or these adjacent channels. This contribution to the signal can thus be subtracted to effectively eliminate optical contamination.

[0095] The characterization of fluorescent probes by determining their respective time signatures, in their unmodified form and their modified form, therefore makes it possible to calculate, at the instant when a thermal cycle is applied to the reaction mixture, the fraction of a modified probe after interaction with the sequence for which it is specific.If this thermal cycle is applied repeatedly during an in vitro nucleic acid amplification reaction, and the fraction of the probe found in the modified form at each cycle is deduced by calculation, this fraction being indicative of the quantity of amplification products resulting from the nucleotide sequence for which the probe is specific, it is possible to establish the curve of the variation of this fraction as a function of time, or of the number of cycles in the case of a PCR-type reaction, and to deduce a threshold cycle (Ct) which then makes it possible to deduce the quantity of the sequence of interest for which the probe is specific (for example by plotting the value of this threshold cycle on a calibration curve).

[0096] In the particular mode where the amplification reaction is the PCR reaction and at least one of the probes is a TaqMan probe, the first component of the temporal signature of this probe consists of the variation in the intensity of its fluorescence during the thermal cycle in the absence of hybridization of this probe with its complementary sequence. The second component of its signature is the variation in the intensity of its fluorescence during the thermal cycle after its hydrolysis, that is to say after cleavage of the fluorophore (or quencher) from the rest of the probe.

[0097] In the particular mode where the amplification reaction is PCR and at least one of the probes used is a molecular beacon, the first component of the temporal signature of this probe consists of the variation in the intensity of its fluorescence during the thermal cycle in the absence of hybridization of this probe with its complementary sequence. The second component of its signature is the variation in the intensity of its fluorescence during the thermal cycle in the presence of its complementary sequence. Advantageously, Molecular Beacon probes have the advantage of providing better fluorescence contrast between their different folding states.

[0098] In the particular mode where the amplification reaction is PCR and at least one of the probes is of the FRET type, the first component of the temporal signature of this probe consists of the variation in the intensity of the fluorescence of the oligonucleotide carrying the acceptor fluorophore during the thermal cycle in the absence of hybridization of this oligonucleotide with its complementary sequence. The second component of its signature is the variation in the intensity of the fluorescence during the thermal cycle in the presence of its complementary sequence, including the FRET signal when the two oligonucleotides of the probe are hybridized adjacently to their common target sequence.Advantageously, since FRET probes can be more flexible in characterizing mutations or in limiting the number of oligonucleotide sequences in the PCR amplification reagent by using several FRET probes on the same amplicon, their use makes it possible to limit the number of oligonucleotides present, and therefore their interaction, and therefore to facilitate the increase in multiplexing.

[0099] Determination of the time signature of a given probe

[0100] For each sequence of interest, a fluorescent oligonucleotide probe is designed comprising an oligonucleotide sequence complementary to and characteristic of a portion of said sequence of interest, and of which the temporal signature (Snm(t), Sm(t)) is known, or failing that, of which the temporal signature is determined, for a given fluorescence channel, comprising two components consisting of the curve of the fluorescence intensity as a function of time for a fixed thermal cycle, when no probe is modified by the interaction with its sequence of interest, and when all the molecules of the probe are modified by the interaction with its sequence of interest respectively.

[0101] This probe can be of the Taqman, Molecular Beacon, dual fluorescence transfer probe, Scorpion, Amplifluor, MGB Eclipse, LUX, QUASR or QZyme type without this list is exhaustive.

[0102] As seen above, for a given fluorescence channel, the time signature comprises a component for the unmodified form of the probe, hereinafter Snm(t), and a component for the modified form of the probe following interaction with its sequence of interest, hereinafter Sm(t).

[0103] The components of the time signature (Snm(t), Sm(t)) can be determined respectively by directly applying the thermal cycle to the probe in the buffer used for the reaction, respectively in its unmodified form and in its form modified by the interaction with its sequence of interest. Alternatively, the time signature can also be obtained by determining the fluorescence profile of the probe in its two forms as a function of temperature (SnmT(T), SmT(T)), and by applying the function obtained to the function describing the temperature profile PT(t) (mathematically, Snm is the composite of the functions SnmT and PT, i.e. SnmToPT; similarly Sm is the composite of the functions SmT and PT, i.e. SmToPT).Alternatively, the components of the time signature of the probe in its unmodified form or in its modified form can be determined indirectly by subtraction between the fluorescence intensity profile of a combination of the probe and other fluorescent probes or sources and the fluorescence intensity profile of this same combination in the absence of the probe whose signature is to be determined. Advantageously, a combination of time signatures can be used rather than the time signature of a probe alone to determine the presence of a probe directly or indirectly.For example, one can use the signature of the combination of several probes and the signature of the combination of these same probes in the absence of one of the probes to determine the absence of this same probe, or the signature of the combination of several probes and the signature of the combination of these same probes in the absence of one of the probes and in the presence of the product of the disappearance of the absent probe to determine the transformation of this same probe (for example the absence of the signature of a TaqMan probe and the presence of the signature of the hydrolyzed probe). The use of a combination of signatures avoids having to characterize each probe individually and makes it possible to overcome possible interaction effects between the probes in the fluorescence signal.In general, the signal of any combination of probes can be used according to the invention as a direct or indirect marker (by combination of several signals) of the state of the composition of the reagent.

[0104] [Fig.3] shows examples of time signature for TaqMan or QUASR type probes using various fluorophores and quenchers: the unmodified time signature component and the modified time signature component following interaction with the sequence for which it is specific.

[0105] [Tables 1] I d Type Fluorophor e(5') Quenc her (3') Sequence 1 TaqMa n FAM BHQ1 FAM-GATACCGCTGGAAACGGCTTTGTCAA-BH Q1 2 TaqMa n ATTO488 BHQ1 ATTO488-CGGCACGGTCAGGTTCGTCCTT-BHQ 1 3 TaqMa n ATTO565 BHQ2 ATTO565-ATTGGCATGGAAGTCACACCTTCGG- BHQ1 4 TaqMa n CY3 BHQ2 CY3-TTCTGACCTGAAGGCTCTGCGCG-BHQ2 5 TaqMa n Texas RED BHQ2 TexasRED-TGCTCGGGCATCATAACGGAAAGC- BHQ2 6 TaqMa n CY5 BHQ2 CY5-GAAGCGCGCGAAATCGAAGTTGCT-BHQ2 7 TaqMa n ATTO647 N BHQ2 ATTO647N-TGCTCGGGCATCATAACGGAAAGC -BHQ2 8 QASR FAM - FAM-ACGCCAATGTTTATGTAAACCTTGCGCC 9 QASR FAM (1) BHQ1( 2) 1: FAM- ACGCCAATGTTTATGTAAACCTTGCGCC 2: GGCGCAAGGTTTACATAAACATTGGCGT- BHQ1 1 0 Mol. FAM BHQ1 FAM-CGGGCACGCCAATGTTTATGTAAACCTT GGCCCG-BHQ1

[0106] [Fig.3] references the “IDs” in Table 1 above and represents for each probe the component of the unmodified time signature (“Av.Int.”) and the component of the modified time signature (“Ap.Int”) following interaction with the sequence for which it is specific.

[0107] According to a preferred embodiment, probes emitting in the same fluorescence wavelength range are chosen so that their temporal signatures (in their unmodified state and in their modified state) are sufficiently distinct to be able to be discriminated.

[0108] Reaction and determination of the concentration of the probes

[0109] Once a set of probes has been chosen, a reaction mix comprising the sample containing the sequences of interest in unknown number and concentration is made with them. In addition to the sequences of interest, this reaction mix comprises, for each sequence of interest sought, the fluorescent probe chosen and whose time signature has been characterized directly or indirectly as indicated above, a set of primers compatible with this probe and capable of amplifying the sequence of interest, reagents and enzymes necessary for carrying out the amplification reaction. In the case of detection by PCR, these include in particular dNTPs and at least one heat-resistant polymerase (in the variant in which the probes are of the TaqMan type, the polymerase has a 5' to 3' exonuclease activity). The reaction mix may also comprise a reverse transcriptase if at least one of the sequences of interest is in RNA.

[0110] In the variant in which the amplification method is PCR and the probes are of the TaqMan or Molecular Beacon type, the primers are non-fluorescent and located on the sequence of interest on either side of the probe. In the sub-variant in which the probes are of the TaqMan type, the polymerase also has a 5' to 3' exonuclease activity.

[0111] In the variant in which the amplification method is PCR and a probe is of the Scorpion, Amplifluor, LUX or QZyme type, this probe also plays the role of one of the two primers.

[0112] Then, the thermocycler 4 subjects the reaction mix to one or more cycles of the temperature profile, this number being able to vary from 1 to 40 typically, if necessary after having carried out a cycle allowing the reverse transcription of sequences of interest into RNA and / or a cycle allowing the activation of the polymerase at high temperature (“hotstart” start). During these cycles the fluorescent probe(s) are excited and the light sensor 6 records the resulting fluorescence continuously on the fluorescence channel(s) corresponding to the fluorescence wavelength ranges of each probe.The sampling frequency of the light sensor 6 is chosen such that at least two fluorescence measurements are carried out per cycle, preferably at least one in the denaturation step and at least one in the hybridization step (which can also be the elongation step in the case of a PCR reaction) preferably at a higher frequency making it possible to cover intermediate temperatures which allows better discrimination power.

[0113] The analyzer 8 is then called to process the fluorescence signals measured in each fluorescence channel by the light sensor 6, by decomposing them, for each channel and at each cycle at which the signals were acquired, in the form of a linear combination of the components of the time signatures of the probes in this channel or a combination thereof. This makes it possible to determine, at each cycle or before and after the appearance of the amplification signal, the fraction of each probe that is in unmodified form, and that which is in modified form. This determination can be carried out qualitatively by visual interpretation of the time signatures or quantitatively by numerical calculation with a suitable algorithm.

[0114] Several algorithms can be used to extract the time signature provided that they allow the decomposition of the measured signatures into a sum of signatures characteristic of the presence of a probe in its modified or unmodified form or of all combinations of presence or absence of several probes in their modified or unmodified form. The algorithm used can for example be an algorithm searching for each cycle, the weights of the linear combination of the characteristic signatures optimally approaching the signatures measured for each cycle. These weights can be directly or indirectly representative of the modification of a probe following its interaction with the sequence for which it is specific, a consequence of the amplification of this sequence by the amplification reaction. The following paragraph details an implementation of this algorithm.

[0115] At the k-th thermal cycle applied during the amplification reaction, and in a given fluorescence channel, the fluorescence signal Fk(t) measured by the light sensor 6 can be defined as follows:

[0116] [Equation = £ "= + £

[0117] Where:

[0118] -1 is the time during the temperature cycle,

[0119] - n is the total number of probes in the reaction mixture,

[0120] - Ci is the quantity (or concentration) of probes of index i (this quantity is known),

[0121] - c™ / { is the quantity (or concentration) of probe of index i present in the form modified at the beginning of the kth cycle of the amplification reaction,

[0122] - represents the time signature of the probe of index i when it is modified, and

[0123] - fnm(tj represents the time signature of the probe of index i when it is non- modified.

[0124] By making the change of variable

[0125] [Equation 2] = Fk(t) - £

[0126] We obtain the equation

[0127] [Equation = £

[0128] Since the quantity n c- fmn(t^ is the sum of the components of the signatures of the unmodified probes for cycle k, and is known since c; is known and independent of cycle k, equation 3 defines a matrix system which associates the measured fluorescence signal with the concentrations of each probe of index i for cycle k.

[0129] These coefficients are those which cancel the matrix equation

[0130] [Equation 4]

[0131] Where the h correspond to the sampling times by the light sensor 6 during the application of the thermal cycle by the thermal cycler 4. It is recalled that the thermal cycle comprises the step of heating from the lower temperature to the higher temperature and the step of cooling from the higher temperature to the lower temperature.

[0132] As the measurements are not perfect, the ch can be chosen for J, K minimize the difference. Many algorithms can achieve this optimization, for example least squares.

[0133] Thus, it appears that, thanks to the time signatures, it is possible to carry out multiplexing / demultiplexing of degree 2, 3, 4, 5 or more, as long as one is able to clearly define the time signatures of each of the probes. This multiplexing / demultiplexing is carried out in all the fluorescence channels of the light sensor 6, which makes it possible to identify the optical contributions of a probe in the channel in which it emits predominantly, but also in the adjacent channels where appropriate.

[0134] The generalized implementation of the algorithm presented above requires characterizing the probes individually, which requires the use of indirect measurements (the signature of the probes when they are not mixed) which can introduce artifacts in the determination of the vector c.

[0135] Advantageously, a similar calculation can be carried out using more direct signatures (for example the signatures of the probes mixed with or without the modification of a probe) by carrying out a change of space by a linear transformation U:

[0136] [Equation 5] >4'0' such that 'W = U .y . .vk(tm),

[0137] such that U allows us to solve this same problem of determining the vector c using more directly measurable time signatures.

[0138] Advantageously, only certain parts of the time signature can be used, to increase the accuracy of the decomposition. Indeed, certain phenomena such as the variation in the quantity of modified probe during the elongation phase of the PCR cycle which results from the activity of the polymerase or the non-reproducibility of the temperature change profile at short times can alter the shape of the signature at certain times and it may be preferable not to take these parts of the signature into account for the determination of the quantities of modified probe. In certain implementation modes, the importance of each time of the time signature can be weighted for the calculation in order to optimize the impact on the result of the calculation.

[0139] Alternatively, the analyzer 8 could operate in a purely digital and non-matrix manner, for example using a gradient descent algorithm to optimize the residual error of the approximation of the signature measured by the linear combination of characteristic signatures, or even contain lookup table type tables, or even use a trained neural network to return the concentrations of each probe as a function of the input measurement signal.

[0140] Advantageously, the algorithm can use a more complex model than the linear combination of signatures, taking into account for example the influence of the variation of the quantity of modified probe within a cycle or the evolution during the cycles of the quantity of modified probes which can be constrained by hypotheses such as the shape of the amplification curve. The algorithm can then take the form of an optimization calculation under constraints by the minimization of an energy function or even take the form of a neural network trained to learn specific parameters of an amplification on the basis of the sequence of signatures.

[0141] [Fig.5] illustrates the implementation in the particular mode of a PCR reaction with two TaqMan probes. It shows the extraction of the respective concentrations of each hydrolyzed probe by decomposition into an optimal sum of the time signatures as described above. The upper curves show the raw fluorescence signal during the PCR with the same mix containing 2 TaqMan probes emitting in the same band, specific respectively to two bacterial genomes which are used pure (the first on the left and the second in the center), or mixed in different quantities (on the right). The lower curves show the respective concentrations of the 2 hydrolyzed probes calculated at each cycle according to an extraction with the same parameters when only the first probe is hydrolyzed during the extraction (left graphs), only the second probe is cleaved (center graphs) and when the 2 probes are cleaved according to different kinetics during the PCR (right graphs). The bottom right graph shows the ability of the device of the invention to extract the amplification signals of these two distinct targets (here sequences belonging to the genome of B. subtilis and E. coli) at different initial concentrations using probes sharing the same fluorescence signal. Examples of implementation:

[0142] Example 1: Identification of the presence of a nucleotide sequence of interest in a multiplex PCR analysis using two TaqMan probes functionalized with fluorophores both emitting predominantly in the band above 650nm.

[0143] The example below shows how the method and device of the invention make it possible to identify the presence of a target sequence of interest among two target sequences using a multiplex PCR analysis using two TaqMan probes functionalized with fluorophores both emitting predominantly in the band above 650nm. These probes, each specific to one of the sequences of interest, emit a fluorescence signal measured in the same channel of the device, but have a different time signature. The use of the demultiplexing algorithm combined with knowledge of these signatures makes it possible to identify the amplification of one target or the other.

[0144] The primers are specific for amplifying two gene sequences [D-alaline—D-alanine ligase]. For E. faecium the sense and antisense primer sequences are GCTTTAGCAACAGCCTATCAG and TCGTCCGAACRTCTTCATTT, for E. faecalis the sense and antisense primer sequences are GTTCTAGTGTCG-GAATTAGCA and GCTTCRATCCCTTGTTCAAC. Two TaqMan fluorescent probes are functionalized at the 5' end with the fluorophore ATTO647N for E. faecalis (sequence TGCTCGGGCATCATAACGGAAAGC, see ID line 7 of Table 1) and CY5 for E. faecium (sequence GAAGCGCGCGAAATC-GAAGTTGCT, see ID line 6 of Table 1). Both probes are functionalized at the 3' end with the quencher BHQ2. We identify the two probes as "Probe a" and "Probe b" respectively.

[0145] In a first step, two PCR reactions are carried out with each primer pair combined with the probe and T DNA of the corresponding bacterial sequence, in order to characterize the signature of each of the two probes separately:

[0146] - in a first PCR, 105 genomes of E. faecalis are amplified (extracted DNA and quantified) and only the E. faecalis-specific probe is added to the PCR mixture at a concentration of 0.1 pM. Both primer pairs are added to the PCR mixture at a concentration of 0.5 pM for each primer. The signal measured in channel 4 of Chronos DX is referred to as “Signal 1”.

[0147] - in a second PCR, 106 genomes of E. faecium are amplified (extracted DNA and quantified) and only the E. faecium probe is added to the PCR mix at a concentration of 0.2pM. Both primer pairs are added to the PCR mix at a concentration of 0.5pM for each primer. The signal measured in channel 4 of Chronos DX is referred to as “Signal 2”.

[0148] In these two PCR reactions, the reaction mixture consists of a buffer (Tris-HCL, 25mM KCl, 16mM (NH4)2SO4, 4.5mM MgCl2) and the amplification is carried out using 3U of Taq DNA Polymerase and 0.2 mM dNTPs. The PCR protocol used consists of: denaturation / activation of the polymerase carried out for one minute at 95°C and 45 PCR cycles (composed of two phases: denaturation for 3 seconds at 95°C and annealing / extension for 15 seconds at 60°C). Fluorescence intensity measurements are carried out every 100ms.

[0149] [Fig.4] shows the signal measured on channel 4 during the two experiments ([Fig.4] a: signal “1” for probe “a” and [Fig.4] b: signal “2” for probe “b”).

[0150] The two components of the time signature of each of the probes are extracted from these signals (signatures shown in lines 6 and 7 of [Fig.3]).

[0151] After defining the signatures of each of these two probes, two multiplex amplification reactions are carried out with a mixture of the reagents using the method of the invention. The two probes are added to the PCR mixture respectively at the concentration of 0.1 pM (probe a, E. faecalis) and 0.2 pM (probe b, E. faecium). The two pairs of primers are added to the PCR mixture at the concentration of 0.5 pM for each primer:

[0152] - in a first PCR, 103 genomes of E. faecalis are amplified (extracted DNA and quantized). The signal measured in channel 4 of Chronos DX is referenced as “Signal 3”. - in a second PCR, 105 E. faecium genomes (extracted and quantified DNA) are amplified. The signal measured in channel 4 of Chronos DX is referred to as “Signal 4”.

[0153] In [Fig.4], the signal measured on channel 4 during these two multiplex PCR reactions is shown ([Fig.4] c: signal “3” and [Fig.4] d: signal “4”).

[0154] Signals “3” and “4” are then demultiplexed using the algorithm of the invention and the time signatures of each probe, by calculating, for each cycle of each multiplex PCR reaction, the representative value of the concentration of each of the modified probes. The result of this demultiplexing is shown in [Fig.4] for each of the multiplex amplification reactions ([Fig.4] e: signal 3 analysis, [Fig.4] b: signal 4 analysis). Observing the curves of the representative values of the concentration of each modified probe, as a function of the PCR reaction cycle, makes it possible to unambiguously identify the nature of the sequence of interest that was ef- effectively present in each of the two reactions.

[0155] Example 2: Identification of the presence of a mixture of two nucleotide sequences of interest in a multiplex PCR analysis using two TaqMan probes functionalized with fluorophores both emitting predominantly in the 575-615nm band.

[0156] In this example, we seek to know the respective quantities of two genomic DNA sequences of two bacteria B. subtilis and E. coli using TaqMan type probes emitting mainly in the same fluorescence band 575-615nm. The E. coli specific probe is labeled with the fluorophore ATTO 565 and the B. subtilis specific probe is labeled with Cy3.

[0157] After determining the time signatures of these two probes as in the example above, a series of 3 PCR amplifications is carried out using a reaction mixture comprising the two pairs of primers and the two probes necessary to amplify the two nucleotide sequences of interest.

[0158] In the first reaction, 104 genomes of B. subtilis were added; in the second reaction, 102 genomes of E. coli were added; in the third reaction, 104 genomes of B. subtilis and 102 genomes of E. coli were added. The reactions were carried out using device 2. The fluorescence signals recorded in channel 3 during each cycle of each of the PCR reactions are shown in the upper row of [Fig.5].

[0159] Algorithm 8 is then used to demultiplex these curves and, for each amplification reaction and for each probe, a curve is obtained showing the representative value of the modified probe as a function of the cycle number. It is observed that, from these demultiplexed curves, the nature of the nucleotide sequence(s) of interest present in the reaction mixtures can be found. For each of these 6 curves, a threshold cycle (Ct) can be measured, which can be plotted in a standard curve that links the Ct cycle and the logarithm of the initial quantity of molecules of each of the sequences of interest in order to quantify the quantity of each of the nucleotide sequences of interest present in the mixture.

[0160] Example 3: Correction of optical contamination caused by the fluorescence signal produced during a PCR during the hydrolysis of a TaqMan probe "a" for the detection of the presence of human and viral sequences of interest

[0161] The PCR experiment is carried out using primers and probes specific to the SARS-CoV-2 viral genome and an endogenous control gene present in the epithelial cells of the nasopharynx in humans. The kit contains probes for the detection of 3 target sequences in 3 optical channels of the device 2: channels 1 and 3 for two targets of the SARS-CoV-2 coronavirus, channel 4 for the endogenous control target. In this experiment, the human DNA contained in a sample is detected by PCR amplification of the control sequence of interest with a pair of primers. The fluorescence signal is generated by a probe functionalized with a fluorophore at the 5' end and a quencher at the 3' end, this fluorophore emits a fluorescence signal in a wavelength band detectable in optical channel number 4 of device 2. This probe is referenced as Probe “a” and its signal as “Signal 1”, shown in [Fig.6] a.

[0162] Due to the spread of the emission spectrum of probe a, the signal generated by the probe is also detected in optical channel number 3 of device 2. This signal is referenced as “Signal 2”. This contamination generates an apparent increase in the signal generated by the probe used for the detection of a Sars-CoV-2 target sequence present in channel number 3, which could be misinterpreted as the presence of this target sequence in the sample. This probe is referenced as probe “b”. Signal 2 is shown in [Fig.6] b.

[0163] The PCR protocol used in this experiment consists of: reverse transcription performed for 30 seconds at 50°C, nucleic acid denaturation and polymerase activation performed for one minute at 95°C, followed by 45 PCR cycles (composed of two phases: denaturation for 3 seconds at 95°C and hybridization / extension for 15 seconds at 60°C). The sampling period of the fluorescence measurement is 100ms for the entire duration of the PCR.

[0164] To eliminate the optical contamination signal, the method according to the invention is applied, comprising the following steps:

[0165] - Determination of the time signature of the 2 probes present in the kit detected in optical channels 3 and 4 of device 2. These signatures are obtained from a PCR with detection of 1.2 μl of the positive amplification control provided with the kit. These signatures are identified as signals 3 and 4.

[0166] - Use of the analyzer algorithm 8 according to the invention to demultiplex the signals present in signal 2. The result of the demultiplexing is shown in [Fig.6] c. It is observed that the curve corresponding to the signal of probe b is perfectly flat, which indicates that the sequence of interest of the SARS-CoV-2 viral genome is absent from the sample.

[0167] Example 4: Detection of a sequence of interest by exploiting the time signature of a fluorescent probe of the “molecular beacon” type used in a multiplex PCR amplification reaction.

[0168] In this example, the method and device of the invention are used to detect the presence of a nucleotide sequence of interest in a sample by means of PCR amplification using “Molecular Beacon” (MB) type probes. In the method using one or more molecular beacons, these probes are not cleaved during the reaction unlike TaqMan type probes. For this purpose, the chosen polymerase lacks 5' to 3' exonuclease activity, and is replaced by strand displacement (SD) activity.

[0169] In this example, a PCR is carried out with the following protocol: initial denaturation / activation of the polymerase for 30 seconds at 92°C and 40 PCR cycles (composed of two phases: denaturation for 5 seconds at 92°C and hybridization / extension for 30 seconds at 62°C).

[0170] The reaction mixture is composed of a buffer (SD polymerase reaction buffer) by adding MgCl2 to the final concentration of 3mM and the amplification is carried out using 5U of SD Polymerase HotStart and 0.2mM of dNTPs. Primers at the concentration of 0.2pM (sense primer, sequence CCGCCAATGGTACCG-CAATCCCT) and 2pM (antisense primer, sequence GCTACTGCCATTA-TATTTTACGGTC) are used to amplify a target sequence of the fimH gene of E. coli (104 bacteria added directly into the PCR after bacterial culture in LB medium and quantification). The fluorescent probe of molecular beacon type (sequence FAM- CGGGCACGCCAATGTTTATGTAAACCTTGGCCCG-BHQ1) is at the concentration of 0.03pM.

[0171] The signal measured in channel 1 of device 2 during the 45-cycle PCR reaction is shown in [Fig.6] d. From this curve, the two components of the time signature of this molecular beacon probe are extracted. These components are shown in [Fig.3] with reference to 1TD 10.

[0172] This signature is then used to demultiplex the signals recorded during multiplex PCR reactions which use this probe to search for the presence of the corresponding nucleotide sequence among other sequences.

[0173] Example 5: Determination of the time signature of a probe according to the QUASR method implemented in a PCR reaction.

[0174] The invention can be used to analyze fluorescence curves during PCR amplification using the QUASR (Quenching of Unincorporated Amplification Signal Reporters) system.

[0175] In this example, a PCR is performed with the following protocol: initial denaturation / polymerase activation for one minute at 95°C and 45 PCR cycles (consisting of two phases: denaturation for 3 seconds at 95°C and hybridization / extension for 15 seconds at 60°C). The sampling period for the fluorescence measurement is 200 ms for the entire duration of the PCR.

[0176] The reaction mixture is composed of a buffer (Tris-HCL, 25mM KCl, 16mM (NH4)2SO4, 4.5mM MgCl2) and the amplification is carried out using 3U of SuperHotTaq DNA polymerase and 0.2mM dNTPs. “Sense” and “antisense” primers of the respective sequence FAM-ACGCCAATGTTTATGTAAACCTTGCGCC and ACATTTCACAA-CACGAGCTGACGA at a concentration of 0.5pM are used to amplify 105 E. coli genomes (extracted and diluted after bacterial culture in LB medium). An oligonucleotide complementary to the “sense” primer GGCGCAAGGTTTACATAAA-CATTGGCGT-BHQ1 is added to the reaction mixture at a concentration of 0.3 pM.

[0177] The fluorescence intensity curve as a function of time is measured during the 45 PCR cycles and illustrated in [Fig.6] e. This curve allows the extraction of the two components of the time signature of the QUASR type probe used. This signature is shown in [Fig.3] with ID 9.

[0178] This signature is then used to demultiplex the signals recorded during multiplex PCR reactions which use this probe to search for the presence of the corresponding nucleotide sequence among other sequences.

[0179] Example 6: Detection of 5 nucleotide sequences in a syndromic kit for respiratory viruses Influenza A, B, respiratory syncytial viruses A and B, and SARS-CoV-2 using a pentaplex PCR.

[0180] The method and device described in the present invention allow the search for the presence of 5 viruses using a pentaplex PCR carried out by mixing five pairs of primers and five probes for the detection of 5 targets using only two optical channels of the device (channels 3 and 4).

[0181] The primers and probes used are specific to the following targets:

[0182] Channel 3:

[0183] Target 1 - Influenza A:

[0184] - sense primer GGAATGGCTAAAGACAAGACCAAT antisense primer CTGCAGTCCTCGCTCACT probe ATTO565-TTCACGCTCACCGTGCCCAGTGA-BHQ2

[0185] Target 2 - Influenza B:

[0186] - sense primer CTCAACTCACTCTTCGAGCGT antisense primer TCTGGTGATAATCGGTGCTCTT probe CY3-TCTGGTGATAATCGGTGCTCTT-BHQ2

[0187] Target 3 - SARS-CoV-2:

[0188] - sense primer GCTTCAGCGTTCTCGGAAT antisense primer CAATTTGATGGCACCTGTGTAG Alexa Fluor probe 568-ATTGGCATGGAAGTCACACCTTCGG-BHQ2

[0189] Channel 4:

[0190] Target 4 - RSVB:

[0191] - forward primer TCCTAACTTCTCAAGTGTGGTC antisense primer CTTGGTTTCTTGGTGTACCTCT probe ATTO647N-AGGCAATGCAGCAGGTCTAGGCAT-BHQ2

[0192] Target 5 - RSV type A:

[0193] - forward primer GCAGGATTGTTTATGAATGCCT antisense primer CACAACTTGTTCCATTTCTGCT probe Cy5-GGTGCAGGGCAAGTGATGTTACGG-BHQ2

[0194] Primers are added to the mixture at a concentration between 0.1 and 0.6pM; probes are added to the reaction at a concentration between 0.1 and 0.5pM. The RT-PCR mixture consists of a buffer (Tris-HCL, 25mM KCl, 16mM (NH4)2SO4, 4.5mM MgCl2) and amplification is performed using between 3 and 10U of TAQ polymerase (with 5' exonuclease activity) and 3U of WarmStart reverse transcriptase. The PCR protocol used consists of: reverse transcription performed for one minute at 60°C, denaturation / activation of the polymerase performed for one minute at 95°C and 45 PCR cycles (composed of two phases: denaturation for 3 seconds at 95°C and annealing / extension for 15 seconds at 60°C). Fluorescence intensity measurements are performed every 100ms.

[0195] 5 PCR reactions were carried out including all the probes except one to determine the time signatures of the probes.

[0196] These time signatures are then used to demultiplex the signals recorded during pentaplex PCR reactions to search, in a single reaction, for the presence of any sequence among the 5 nucleotide sequences of interest.

[0197] Thus, it appears that, thanks to the time signatures, it is possible to carry out multiplexing / demultiplexing of degree 2, 3, 4, 5 or more, as long as one is able to properly define the time signatures of each of the probes.

[0198] This capacity for multiplexed detection of nucleic acid sequences opens up significant possibilities, such as, for example, the possibility of carrying out the simultaneous search for numerous point genetic mutations of the genome of the SARS-CoV-2 coronavirus in a clinical sample, which makes it possible to rapidly identify the variant present in this sample, a faster and less expensive alternative than the sequencing of the viral genome present in this sample.

Claims

Claims

1. A device for multiplexed detection of nucleic acid sequences comprising a thermal cycler (4) arranged to perform a series of thermal cycles with an in vitro nucleic acid amplification reagent containing at least two fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores emitting in overlapping fluorescence wavelength ranges, a light sensor (6) arranged to measure radiation emitted by said fluorophores in said fluorescence wavelength ranges, the radiation emitted by each probe varying depending on whether it is in an unmodified state in which the quencher attenuates or does not substantially attenuate the fluorescence emission, or in a modified state in which the quencher has an opposite effect on the fluorescence emission,and an analyzer (8) arranged to determine, for each respective fluorescence probe, a value representative of a concentration in a modified state, from time signatures derived from the fluorescence measured as a function of time for a given thermal cycle of a reaction mixture comprising one or more probes, which may each be substantially entirely in an unmodified or modified state, and from at least two measurements carried out during at least some of the thermal cycles, which values representative of a concentration in a modified state make it possible to qualify the presence of one or more nucleic acid sequences each associated with a distinct fluorescence probe so as to cause a change of state of the fluorescence probe from the unmodified state to the modified state when they interact.,

2. The device of claim 1, wherein the thermal cycler (4) is a PCR thermal cycler, and the at least two fluorescence probes are selected from the group consisting of TaqMan probes, molecular beacons, dual fluorescence transfer probes, Scorpion probes, Amplifluor probes, MGB Eclipse probes, LUX probes, QUASR probes and QZyme probes.

3. A device according to claim 1 or 2, wherein the in vitro nucleic acid amplification reagent comprises three fluorescence probes combining a quencher and a fluorophore, said probes fluorescence being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges.

4. A device according to any preceding claim, wherein the in vitro nucleic acid amplification reagent comprises at least two groups of fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges within each group of fluorescence probes.

5. Device according to one of the preceding claims, wherein the analyzer (8) is arranged to determine the representative value of the concentration of each fluorescence probe in a modified state by solving a matrix system based on the fluorescence measurements, the time signatures and the constancy of the respective concentration of each fluorescence probe.

6. Device according to claims 4 and 5, wherein the analyzer (8) is arranged to resolve a matrix system by group of fluorescence probes.

7. Device according to claim 5 or 6, wherein the analyzer (8) is arranged to use a minimization algorithm.

8. Device according to one of claims 1 to 4, in which the analyzer (8) is arranged to determine the representative value of the concentration of each fluorescence probe in a modified state by applying a gradient descent or by using a neural network.

9. A method for multiplexed detection of nucleic acid sequences comprising the following operations: a) performing a plurality of thermal cycles on a reaction mixture comprising a sample to be analyzed and an in vitro nucleic acid amplification reagent containing at least two fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores emitting in overlapping fluorescence wavelength ranges, b) during each thermal cycle, performing at least two fluorescence measurements in said fluorescence wavelength ranges, the radiation emitted by each probe varying depending on whether it is in an unmodified state in which the quencher attenuates or does not attenuate the fluorescence emission substantially, or in a modified state in which the quencher has an opposite effect on the fluorescence emission, c) determining a representative value of the concentration of each fluorescence probe in a modified state from the measurements of step b) and time signatures derived from the fluorescence measured as a function of time for a given thermal cycle of a reaction mixture comprising one or more probes, each of which may be substantially entirely in an unmodified or modified state, which representative values of a concentration in a modified state make it possible to qualify the presence of one or more nucleic acid sequences each associated with a distinct fluorescence probe so as to cause a change of state of the fluorescence probe from the unmodified state to the modified state when they interact.

10. The method of claim 9, wherein step b) comprises performing PCR cycles, and the at least two fluorescence probes are selected from the group comprising TaqMan probes, molecular beacons, dual fluorescence transfer probes, Scorpion probes, Amplifluor probes, MGB Eclipse probes, LUX probes, QUASR probes and QZyme probes.

11. A method according to claim 9 or 10, wherein step b) comprises using an in vitro nucleic acid amplification reagent comprising three fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges.

12. A method according to any one of claims 9 to 11, wherein step b) comprises using an in vitro nucleic acid amplification reagent comprising at least two groups of fluorescence probes combining a quencher and a fluorophore, said fluorescence probes being arranged to each target a distinct nucleic acid sequence and said fluorophores having overlapping fluorescence wavelength ranges within each group of fluorescence probes.

13. Method according to one of claims 9 to 12, in which operation c) includes the resolution of a matrix system based on fluorescence measurements, time signatures and the constancy of the respective concentration of each fluorescence probe.

14. A method according to claims 12 and 13, wherein step c) comprises solving a matrix system by group of fluorescence probes.

15. Method according to one of claims 9 to 13, in which operation c) comprises the application of gradient descent or the use of a neural network.

Images