Method and system for controlling preparatory liquid chromatography

Description

FIELD OF THE INVENTION

[0001] The present invention relates to a method and a system for controlling preparative liquid chromatography, from low to high pressure. STATE OF THE ART

[0002] Preparative liquid chromatography is a process for isolating, purifying or enriching a mixture of compounds which, to be optimal, requires the selection of a certain number of parameters regarding the separation tool used (inner diameter, length, stationary phase, particle size, etc.) and the operating conditions (mobile phase, elution type, flow rate, temperature, injection mode, injected quantity, etc.). The choice of these parameters often involves a productivity compromise between the desired purity level, the desired quantity of pure product, the cost and the duration of the purification process.

[0003] To make this choice, it is known to use separation results obtained, for a sample of the same nature, with an analytical chromatographic system, and to transpose them to preparative liquid chromatography. The Anglo-Saxon term "scale-up" thus defines a multiplying coefficient between the quantity of sample treated in analytical liquid chromatography and the quantity of sample treated in preparative liquid chromatography.

[0004] However, the analytical liquid chromatography system is quite different from that of preparative liquid chromatography in terms of stationary phase and operating conditions, so this transposition is relatively complex. Calculation tools have therefore been designed to facilitate this transposition.

[0005] Thus, document US 7,686,959 describes a computer-implemented method using thin layer chromatography (TLC) data to calculate operating conditions for preparative chromatography. More specifically, this document teaches the calculation, from thin layer chromatography, of the frontal ratio R f1 , R f2 for two compounds of the sample to be purified, the calculation of column volumes CV 1 , CV 2 for each of said compounds from their respective frontal ratio (each column volume being equal to the inverse of the respective frontal ratio), and the calculation of a column factor ΔCV defined as the difference between the volumes CV 2 and CV 1 . However, this factor ΔCV is specific to each column, which requires a large number of calculations to determine the optimal column.Furthermore, this process requires a large number of user interventions (data entries), which assumes that the user has solid knowledge of chromatography.

[0006] Document US6413431 describes a method in which analytical HPLC or TLC chromatography is carried out, and where operating conditions for preparative chromatography are determined from the analytical chromatography data.

[0007] However, since an operator responsible for carrying out purification in preparative liquid chromatography is not necessarily a chromatography specialist, it would be useful to offer a process and a control system which is more automated and which requires only minimal intervention from the operator. STATEMENT OF THE INVENTION

[0008] The aim of the invention is therefore to design a method for controlling preparative liquid chromatography which can be implemented by a user who is new to chromatography, minimizing the amount of data to be entered manually to determine the separation tool and the optimal and sufficient purification conditions depending on the expected result.

[0009] The invention is defined in the claims.

[0010] To this end, the invention proposes a method for controlling preparative liquid chromatography, comprising the following steps, at least part of said steps being implemented by a computer comprising a processor and a display screen coupled to said processor: (a) selecting, via an interface, an analytical liquid chromatography method from thin layer chromatography (TLC) and high performance liquid chromatography (HPLC), (b) entering analytical liquid chromatography data obtained by the method selected in step (a) for a product to be purified, (c) accessing a table of separation tools available to the user to carry out said preparative liquid chromatography, said tools being columns, cartridges or other chromatography devices comprising a stationary phase, (d) from said analytical liquid chromatography data and the table of available separation tools, selecting an optimal separation tool from said table and calculating preparative liquid chromatography operating conditions on said selected separation tool.This step includes in particular the determination of a suitable gradient profile.

[0011] The method further comprises a step of receiving the quantity of product to be purified.

[0012] According to one embodiment, the method further comprises a step of selecting the mode of introduction of the product to be purified onto the selected separation tool, said mode of introduction being chosen from an introduction in solution or a solid deposition.

[0013] Advantageously, the method also comprises a step of validation of the calculation by a user, and after said validation, a step of launching the preparative liquid chromatography on the selected separation tool.

[0014] According to one embodiment, the method comprises the activation by a user of an acceleration of the preparative liquid chromatography. Step (d) then comprises the calculation of the optimized operating conditions taking into account the maximum flow rate and the limiting pressure of the preparative liquid chromatography system in which the selected separation tool is implemented.

[0015] According to one embodiment, the method selected in step (a) is thin layer chromatography (TLC). Step (b) then comprises entering analytical data obtained on a thin layer chromatography plate.

[0016] Preferably, the selection and calculation implemented in step (d) take into account four different cases depending on the retention factor values ​​Rf of the plate: 0 , 01 < Rf < 0 , 07 , 0 , 08 < Rf < 0 , 4 , 0 , 4 < Rf < 0 , 9 , And 0 , 9 < Rf < 1 , 0

[0017] Particularly advantageously, the method comprises an adaptation of the shape of the elution gradient to each value of a volume difference factor ΔCV taking into account the fact that the same factor ΔCV corresponds to different retention factor differences ΔRf in each of said four zones.

[0018] Preferably, the selection and calculation implemented in step (d) take into account from one to all of the compounds of interest distributed over the entire plate.

[0019] Advantageously, step (d) takes into account the difference between the eluent strengths of the solvents used in the preparative liquid chromatography and the non-linear variation of the eluent strength of a mixture of solvents as a function of the composition of said mixture.

[0020] According to one embodiment, the method comprises acquiring an image of a thin layer chromatography plate and automatically detecting, from said image, the separated compounds by a handheld device of the user, the input of the analytical data of step (b) comprising an import of data from said handheld device.

[0021] According to one embodiment, the method selected in step (a) is high performance liquid chromatography, the selection further comprising the choice of a chromatography mode from: normal phase chromatography (NPLC), reversed phase chromatography (RPLC), hydrophilic interaction chromatography (HILIC), hydrophobic interaction chromatography (HIC).

[0022] The analytical chromatography data entered in step (b) are then analytical liquid chromatography data obtained on a reference separation tool.

[0023] According to one embodiment, the table of available separation tools comprises a separation tool identical to the reference separation tool, and step (d) comprises the selection of said tool and the calculation of the preparative liquid chromatography operating conditions for said separation tool.

[0024] Alternatively, the table of available separation tools does not include a separation tool identical to the reference separation tool, and step (c) comprises the selection of a separation tool having a stationary phase different from that of the reference tool by comparing similarities between the stationary phase of the reference tool with the stationary phase of the available separation tools, or the selection of a separation tool comprising the same stationary phase as the reference tool and the calculation of the operating conditions for preparative liquid chromatography in multi-injection mode (“multi-run” according to English terminology) by minimizing the number of injections.

[0025] According to one embodiment, the method comprises a step of calculating separation by thin layer chromatography or by high performance liquid chromatography at isoelution to determine other analytical conditions.

[0026] Advantageously, the selection and calculation in step (d) takes into account the eluent strength specific to each strong solvent.

[0027] According to one embodiment, the method comprises implementing a separation test ("screening" according to English terminology) on a plurality of separation tools and selecting the separation tool having the best separation potential from among said tools. This test is implemented when there is no analytical data capable of being entered in step (b), whether in thin layer chromatography or in high performance liquid chromatography.

[0028] According to one embodiment, the table of available separation tools comprises a separation tool identical to the tool having the best separation potential, and step (d) comprises the selection of said tool and the calculation of the preparative liquid chromatography operating conditions for said separation tool.

[0029] Alternatively, the table of available separation tools does not include a separation tool identical to the tool with the best separation potential, and step (d) comprises selecting a separation tool comprising the same stationary phase as the tool with the best separation potential and calculating the operating conditions for preparative liquid chromatography in multi-injection mode by minimizing the number of injections.

[0030] According to one embodiment of the invention, the method comprises a sequence of at least two series of steps (a) to (d), with two different analytical liquid chromatography methods and / or modes.

[0031] Particularly advantageously, said sequence is carried out automatically when a first series of steps (a) to (d) has not made it possible to identify a preparative liquid chromatography solution.

[0032] Another object of the invention is a system for implementing the method which has just been described.

[0033] The said system includes: a computer comprising a processor, a user interface and a display screen, the processor being configured to implement the steps of the method as described above, a preparative liquid chromatography system controlled by said computer, comprising a mobile phase reservoir, a pump, an injector, a separation tool, a detector, a data logger and a fraction collector.

[0034] Preferably, the control method described above will be implemented by means of a processing device comprising means for implementing the steps of the control method, such as a PC type computer comprising a memory and a processing unit on which a computer program is executed.

[0035] This computer program includes in particular one or more algorithms making it possible to execute the steps of the method described above. DESCRIPTION OF FIGURES

[0036] Other characteristics and advantages of the invention will emerge from the detailed description which follows, with reference to the appended drawings in which: there figure 1 is a flowchart showing the operation of the system according to the invention; the figure 2 is a diagram of a separation obtained on a thin-layer chromatography plate. DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

[0037] The invention makes it possible, from an analytical chromatographic analysis, in particular on a thin layer or on an HPLC column (high-performance liquid chromatography, acronym for the English term "high pressure liquid chromatography" or "high performance liquid chromatography"), in normal phase, reverse phase or other, to propose the best possible compromise of purification by preparative liquid chromatography.

[0038] Said preparative liquid chromatography can be implemented from low pressure to high pressure, and includes in particular the names flash chromatography, MPLC (acronym for the English term "medium pressure liquid chromatography"), and HPLC. The implementation pressure is limited only by the chromatography system used, and in particular the pressure that the separation tool can withstand. By "low pressure" we generally mean a pressure of the order of a few bar to several tens of bar; by "high pressure" we generally mean a pressure of a few tens of bar to several hundred bar, or even more.

[0039] To this end, the invention provides a computer-implemented method that guides the user from providing analytical data to defining all the conditions that enable preparative liquid chromatography to be carried out. All of the steps can be carried out automatically, without the user being prompted to enter data or make choices.

[0040] This process is therefore particularly interesting for a user who is new to chromatography.

[0041] However, even when the user has a certain level of expertise in chromatography, the method remains advantageous insofar as it allows the time required to determine the operating conditions to be reduced; moreover, the method can include steps implemented directly by the user, which allow the latter to interact with the algorithms used (expert mode).

[0042] The method includes access to a table of separation tools available at the user's disposal. This table may be presented in the form of a database listing all the tools in stock at the user's disposal, and comprising fields corresponding to the main characteristics of said tools: type of tool (for example: column, cartridge), material constituting the tool (plastic, stainless steel, etc.), dimensions (internal diameter, width, etc.), nature of the stationary phase, particle size of the stationary phase, brand or supplier of the stationary phase, etc. In the remainder of the text, the term "column" will be used for the sake of brevity, but without any limitation as to the type of separation tool used.Particularly advantageously, said preparation tools make it possible to implement different chromatography techniques, in particular techniques which can be implemented at different pressures (from low pressure to high pressure), in normal phase and / or in reverse phase.

[0043] The method can advantageously be implemented in the chromatography systems marketed by the Applicant under the brand name puriFlash ™<, but could also be implemented in other systems. These systems generally comprise a mobile phase reservoir, a pump, an injector, a separation tool, a detector, a data logger and a fraction collector. Such systems are well known to those skilled in the art and will therefore not be described in detail in this text.

[0044] The method may include access to a table of chromatography systems (or equipment) available to the user. The method takes into account the needs expressed by the user according to the stage of product development, for example in terms of purity to be achieved, to propose the most suitable chromatography system. The definition of the system takes into account the dead volumes of the equipment and the necessary pumping power. Thus, the method makes it possible to choose the most suitable tool and chromatography system depending on the product to be purified and the desired degree of purity.

[0045] The control unit comprises a processor configured to implement a number of algorithms described below, or to communicate with a remote computer implementing said algorithms. Optionally, in the case of a solid sample, the system may also comprise a solid injection support (generally called “dry load” according to English terminology); this support is arranged on the mobile phase circuit upstream of the column.

[0046] The flowchart of the figure 1 illustrates the general architecture of the process.

[0047] The different methods are described successively below.

[0048] It is interesting to note that all the modalities are linked together, so that they can all possibly be implemented, starting from TLC analytical data which are the data most classically available to the user, going towards other types of analytical data when the TLC analytical data are not available or are not satisfactory. Thus, when one type of analytical data does not allow the user to be offered a preparative liquid chromatography solution, the user is automatically guided towards another type of analytical data likely to lead to the proposal of a separative liquid chromatography solution. Consequently, even if he is a novice in chromatography, the user is fully guided from the provision of analytical data to obtaining a preparative liquid chromatography solution.

[0049] Naturally, depending on the nature of the sample (polar or not) and the nature of the analytical data available, only some of these steps can be implemented, the user being free to launch the implementation of the process from any of the said modalities. However, the system is configured to be able to process all the types of analytical data detailed below.

[0050] In a first step, the user is prompted to select, by means of an interface, the analytical liquid chromatography process used to obtain the analytical data to be used. This selection is made from among the following processes, which are proposed to the user by the interface: thin layer chromatography (TLC), high performance liquid chromatography (HPLC).

[0051] Where applicable, the selection also includes the chromatography mode used to obtain the HPLC column data. The chromatography mode is typically chosen from: normal phase chromatography (NPLC), reverse phase chromatography (RPLC), hydrophilic interaction chromatography (HILIC, acronym for the English term "hydrophilic interaction chromatography"): this technique is particularly advantageous for the separation of small polar molecules, hydrophobic interaction chromatography (HIC, acronym for the English term "hydrophobic interaction chromatography"): this technique is particularly advantageous for the separation of hydrophobic amino acids, peptides and proteins.

[0052] In a second step, the analytical data obtained by the previously selected method are provided to the processor as input data, via an interface. As will be described in detail below, these data can be entered manually by the user into a table, or imported from a portable device such as a smartphone or tablet in which the data are stored.

[0053] A first series (P) of situations concerns a polar sample.

[0054] According to a first possibility (box 1), the user has analytical data on a TLC plate, for the sample considered.

[0055] The processing of CCM analytical data was the subject of an original development by the inventors.

[0056] First, we recall the definition of the quantities used: the retention factor Rf is a relative quantity that characterizes the elution of a compound on a thin layer chromatography (TLC) plate. It is the ratio between the migration distance of a solute and the migration distance of the mobile phase. Thus, a completely eluted solute has a retention factor equal to 1. A solute that is not entrained by the mobile phase has a retention factor equal to 0. k is a thermodynamic quantity characterizing the retention of a solute in a column, which is equal to (tr-t0) / t0 or (Vr-V0) / V0, where tr is the solute retention time, t0 the time taken for the mobile phase to travel through the column, Vr the volume of mobile phase required to remove the solute from the column, V0 the void volume of the column.

[0057] For the same chromatographic system, there is a fundamental relationship linking the retention factors Rf and k: k = 1 − Rf / Rf Δk is a relative quantity independent of the geometry of the column, which is characteristic of a given solute, eluted on a given stationary phase by a given mobile phase at a given temperature. In preparative liquid chromatography, it is the volume of mobile phase necessary to collect the solute which is the really useful quantity. This volume Vs, sometimes called CV, is expressed in units of dead volume of the column to take into account the different column geometries: Vs = CV = Vr / V 0 = 1 + k

[0058] There figure 2schematically illustrates a TLC plate with two spots each corresponding to a compound a, b, one of which is the compound of interest and the other its nearest neighbor. ds is the distance between the deposition line (designated by the letter D) of the sample and the solvent front (designated by the letter F). da is the distance between the deposition line of the sample and the center of the spot corresponding to compound a, and db is the distance between the deposition line of the sample and the center of the spot corresponding to compound b. Rf a = d a / d s Rf b = d b / d s Vs a = CV a = 1 / Rf a = 1 + k a Vs b = CV b = 1 / Rf b = 1 + k b from where: ΔCV = CV b − CV a k = K tr × 1 / Rf − 1 avec K tr = constante = 1 Δk = K tr × 1 / Rf b − 1 − 1 / Rf a − 1 from where: ΔCV = Δk

[0059] The invention can be implemented with a TLC plate whose retention factor Rf of the compound of interest is between 0.01 and 1.0 with a retention difference Δk between the compound of interest and its nearest neighbor (or its two nearest neighbors) greater than or equal to 0.20 and for ΔRf ≥ 0.02.

[0060] It can take into account from one to all of the compounds present on the TLC plate, said compounds being able to be distributed over the entire TLC plate. Thus, a transposition of thin layer chromatography to preparative liquid chromatography can always be implemented, which makes it possible to offer the user a purification solution without requiring them to carry out additional manipulations.

[0061] The originality of the treatment is that it takes into account the amplitude Δε° (i.e. the difference between the eluting forces of two solvents taken separately) and the non-linear variation of the eluting force of a mixture of pure solvents as a function of the composition of the mixture, ε° being the eluting force of a solvent.

[0062] The calculations define five working amplitudes representative of all the amplitudes covering the eluting forces of the solvents entered in the software.

[0063] In addition, the inventors defined, based on the absolute and relative position of the compounds on the TLC plate, four retention zones of the compound of interest.

[0064] For example, for the cyclohexane / ethyl acetate couple, for which Δε° = 0.38, the following zones were defined: 0 , 01 < Rf ≤ 0 , 07 − 0 , 08 ≤ Rf ≤ 0 , 40 − 0 , 40 < Rf ≤ 0 , 9 − 0 , 9 < Rf < 1 , 0 and, for each of these zones, the following six strong solvent conditions were defined to piecewise mimic the curvature of the eluting force (strong solvent being the most eluting solvent in the mobile phase): where % S strong is the volume percentage of strong solvent in the mobile phase.

[0065] For each of these retention zones, the calculation of the Δk of the critical pair or triplet (i.e. the pair or triplet of peaks which are the least well resolved) allows, knowing the void volume V0 of the columns, to propose a certain number of gradient conditions and isocratic conditions and as a function of the size of the particles in the column (box 4).

[0066] This leads to 1215 cases in total (for 115 of these cases, separation is not possible).

[0067] The shape of the elution gradient is thus adapted to each value of ΔCV taking into account the fact that the same ΔCV factor corresponds to different differences in retention factor ΔRf in each of the four retention zones mentioned above.

[0068] From these elements, a computer algorithm was built to chain all the conditions relating to the four zones and lead to a single preparative liquid chromatography solution. This algorithm integrates a direct multiplication factor calculation ("scale-up") based on the sizing of the column to be used according to the quantity of sample to be purified. Said factor is the multiplication factor between the quantity of material in analytical chromatography and that in preparative liquid chromatography.

[0069] The user specifies, as additional input data, the mass of sample to be purified.

[0070] The user can also select the method of introducing the product to be purified, between an introduction in solution or a solid deposit (“dry load”).

[0071] An algorithm calculates the best match between the geometry and nature of the chromatography column, based on the loading equations to improve the cost / productivity ratio of preparative liquid chromatography.

[0072] The initial conditions and the shape of the gradients are calculated from the Rf and ΔRf of the solutes in TLC. The different gradient profiles proposed are based on the shape of the eluent force curves of binary solvent mixtures in HPLC with normal phase polarity (normal phase mode).

[0073] A classification of six columns is ordered according to the separation difficulty Δk and ΔRf (box 6). This classification is made from the column table, so that the proposal contains at least three columns in stock at the user's disposal. This classification takes into account the column type, the stationary phase type and the molecular mass of the compounds of interest.

[0074] The display of the columns offered among all available columns is therefore modulated according to the targeted application and the input data.

[0075] The user is invited to validate the chosen column (box 7).

[0076] There are different ways to enter analytical data.

[0077] In one embodiment, the user manually enters analytical data into a table to be completed (box 2). The data entered typically includes: for TLC: the Rf, the compounds of interest, the solvents and the mass to be purified; for HPLC: the retention times of the compounds, the sum of the dead volumes upstream of the column (known as "dwell volume," in English), the internal diameter of the column and its length, the particle size of the stationary phase, the dead volume of the analytical column, the flow rate and the injected volume.

[0078] According to a particularly advantageous embodiment, the user has a smart portable device, such as a smartphone or a tablet, equipped with an application dedicated to the acquisition of TLC analytical data and the transfer of said data to the processor intended for the preparation of a preparative liquid chromatography solution (case 3). Such an application has the advantages of making the thin layer chromatography step and its processing faster and simpler.

[0079] The said application has the following features: (1) automatic detection of compounds on the TLC plate:

[0080] From a photograph of a TLC plate taken live by the user or downloaded from the photo library located in the user's smartphone, the algorithm distinguishes (if the contrast allows) the outline of the shapes and places a point in their center indicating the presence of a compound. If the compound is not automatically recognized by the algorithm, the user can manually identify the compounds or make corrections. He must also place two lines representing the sample deposition line and the solvent front. (2) identification of compounds of interest and calculation of retention factors (Rf) and ΔCV (=Δk)

[0081] Once all the compounds have been identified, the user indicates those of interest by clicking on them. The retention factors Rf are displayed for each compound of interest. If desired, by pressing a dedicated button, the user can see if their compounds are within the working comfort zone. The zone is typically between 0.1 and 0.4.

[0082] The user can also, by clicking on a dedicated button, display the ΔCV (=Δk). These are calculated from the placement of the upper and lower compound.

[0083] The application communicates the lowest ΔCV (=Δk) in the form "Min ΔCV". From this data, an alert message is displayed indicating to the user the difficulty level of the separation. For example, the manual mode can include three difficulty levels: easy, corresponding to a “min ΔCV” greater than 4, standard, corresponding to a “min ΔCV” between 1.5 and 4, difficult, corresponding to a “min ΔCV” less than 1.5. (3) information on solvents and the quantity of sample to be purified:

[0084] The user is asked to enter the quantity of sample to be purified and to select the solvents (maximum two) that were used and their respective proportion in %. For each solvent, the user can indicate whether there is an additive and add a comment about them or more generally on the next screen if desired.

[0085] A summary screen of your data is then displayed. (4) the direct and secure transmission of this information to the chromatography system:

[0086] The above-mentioned data can be saved in the user's personal database, sent to the user by email, or sent via Bluetooth or Wifi its information to the preparative liquid chromatography system. (5) pairing the application to the preparative liquid chromatography system:

[0087] The application includes a configuration module that allows the chromatography system to be paired with the application via Wi-Fi, using a file transfer protocol (FTP) where the analytical data file will be stored and read. This configuration module can also allow the user to enter their email address to receive the files on it. (6) archiving and repeated sending of the same data:

[0088] The application advantageously includes an archiving module in which the user can, if desired, archive their data. In this case, the data will only be viewable from their smartphone. If necessary, the user can resend the analytical data from a TLC plate to the chromatography system as many times as desired.

[0089] Advantageously, the sending is done via a .json file including a photograph of the plate, the name assigned to it, the Rf of the compounds of interest, comments on the plate in general, the indication of the quantity of product to be purified, the name of the solvents, their proportion in %, the mention of an additive and any comments on the solvents if applicable. All this data is read by the processing algorithm described below in order to develop proposals for preparative liquid chromatography solutions.

[0090] However, it may happen that the TLC separation is not satisfactory, for example due to the positioning of the peak(s) of interest, the fact that the nature or quantity of solvent to carry out the purification is greater than the user's stock, the fact that no suitable column is in stock, etc.

[0091] In this case, the processor launches an algorithm which, by changing the analytical conditions, in particular the strong solvent or the pair of solvents used, makes it possible to obtain, at iso-elution, separations different from the initial ones (box 5). This makes it possible to determine the experimental conditions with the best potential in the optimum retention zone to obtain efficient purification. The algorithm is based for this purpose on the universal diagram described in the book Principles of adsorption chromatography, RL Snyder, ed. M. Dekker (1968).

[0092] The algorithm calculates specific strong solvent percentage limits for each of the five working ranges mentioned above.

[0093] Then, preparative liquid chromatography conditions are proposed (box 6 described previously).

[0094] In the absence of TLC analytical data, but if an analytical chromatogram is available (box 9), another module is used, as described below.

[0095] In the case of a chromatogram obtained by HPLC, the resolution between two peaks is defined by the relation: Rs = 2 × t R 2 - t R 1 / ω 2 − ω 1 , where t R is the retention time of a peak, and ω is the width of a peak at its base.

[0096] It follows that the separation between two peaks is all the better as Rs is large.

[0097] In the first case (box 10), the analytical stationary phase is identical to that of at least one column in the column table. From the analytical column entered, an algorithm identifies one or more columns in the column table whose fields relating to the stationary phase material, its brand and its grafting are in agreement. The quantity of sample to be purified is entered.

[0098] Another algorithm (box 11) selects the column(s) compatible with the system in terms of pressure, taking into account the nature of the stationary phase, the particle size, the nature of the solvents and the column material.

[0099] An algorithm calculates the scale-up ratio from the internal diameter and length of the columns and the percentage of pore volume of the stationary phase particles used.

[0100] The mass of raw sample to be purified is divided by the said scale-up ratio to calculate the number of injections to be performed.

[0101] Finally, an algorithm performs the transposition of the gradient method of an analytical column to a gradient method of a preparative liquid chromatography column. This transposition is based on the direct ratio of the column dimensions, the recalculation of the gradient slope, its steps and times and the percentage of pore volume of the particles used. It also takes into account the void volume V0 of the columns and the delay volume of the apparatus. It thus predicts the evolution of the pressure and the solvent consumption. The retention times of the compounds in analytical entered in a retention time calculation table are similarly transposed into predictive retention times on the preparative liquid chromatography column.

[0102] In the case where the user does not have a column with the same stationary phase as that of the analytical evaluation (box 12), an algorithm identifies in the column table whether the field relating to the nature of the stationary phase corresponds to (a) a virgin silica, (b) a grafted silica other than C18 or (c) a C18 grafted silica.

[0103] In cases (a) and (b) (silica or grafted silica other than C18), the algorithm identifies whether the fields relating to the stationary phase and the grafting type in the column table are correlatable; if so, the algorithm proposes a direct transfer.

[0104] In case (c) (C18 grafted silica, which can present large differences in selectivity depending on the grafting carried out), a table, called "Atom table", compiling the experimental data of several hundred columns based on C18 grafted silica, from different brands and manufacturers, was previously constructed. The algorithm performs a normalization of the experimental values ​​according to the extremes. For each analytical column and for each of the qualification criteria, a Euclidean distance relative to all the other columns is calculated. After normalization, each distance of each criterion is added to become the reference value of this column with respect to all the others. One of the criteria consisted of defining a unified parameter relating to hydrophobicity.

[0105] The algorithm searches in the said Atom table for the stationary phase, brand and grafting type fields identical to those of the analytical column entered (box 13).

[0106] If it recognizes it, the algorithm calculates the proximities (Euclidean distance) of the selectivities of all the stationary phases that are in the Atom table. It classifies the responses in ascending order and only retains candidates whose proximity values ​​are less than 0.17.

[0107] The algorithm compares its results to the Atom table and retains only six candidates according to criteria based on the number of injections and their presence in the column table, which reflects the fact that they are in physical stock at the user's premises.

[0108] The following steps are those described for box 11.

[0109] If the user does not have a suitable column (box 14), they are notified that they must purchase a suitable column. Depending on the column purchased, the following steps are those described for box 11.

[0110] If no direct normal phase solution is found (box 15), a normal phase screening is implemented (box 16). Such screening consists of studying the separation on different normal phase selectivities in an analytical column from an HPLC or flash chromatography system.

[0111] Following this screening, an algorithm selects the column with the best purification potential (box 17). An algorithm can possibly propose elution conditions allowing different separations to be obtained.

[0112] The algorithm then uses an analytical chromatogram of the separation with the selected column (box 18).

[0113] In the first case (box 19), the analytical stationary phase is identical to that of at least one column in the column table. From the analytical column entered, an algorithm identifies one or more columns in the column table whose fields relating to the stationary phase material, its brand and its grafting are in agreement. The quantity of sample to be purified is entered.

[0114] Another algorithm (box 20) selects the column(s) compatible with the system in terms of pressure, taking into account the nature of the stationary phase, the particle size, the nature of the solvents and the column material.

[0115] An algorithm calculates the scale-up ratio from the internal diameter and length of the columns and the percentage of pore volume of the stationary phase particles used.

[0116] The mass of raw sample to be purified is divided by the said scale-up ratio to calculate the number of injections to be performed.

[0117] Finally, an algorithm performs the transposition of the gradient method of an analytical column to a gradient method of a preparative liquid chromatography column. This transposition is based on the direct ratio of the column dimensions, the recalculation of the gradient slope, its steps and times and the percentage of pore volume of the particles used. It also takes into account the void volume V0 of the columns. It thus predicts the evolution of the pressure and the solvent consumption. The retention times of the analytical compounds entered in a retention time calculation table are similarly transposed into predictive retention times on the preparative liquid chromatography column.

[0118] In the case where the user does not have a column with the same stationary phase as that of the analytical evaluation (box 21), an algorithm identifies in the column table whether the field relating to the nature of the stationary phase corresponds to (a) a virgin silica, (b) a grafted silica other than C18 or (c) a C18 grafted silica (box 22).

[0119] In cases (a) and (b) (silica or grafted silica other than C18), the algorithm identifies whether the fields relating to the stationary phase and the grafting type in the column table are correlatable; if so, the algorithm proposes a direct transfer (box 20).

[0120] In case (c) (C18 grafted silica, which can present large differences in functionality depending on the grafting carried out), a table, called "Atom table", compiling the experimental data of several hundred columns based on C18 grafted silica, from different brands and manufacturers, was previously constructed. The algorithm performs a normalization of the experimental values ​​according to the extremes. For each analytical column and for each of the qualification criteria, a Euclidean distance relative to all the other columns is calculated. After normalization, each distance of each criterion is added to become the reference value of this column with respect to all the others. One of the criteria consisted of defining a unified parameter relating to hydrophobicity.

[0121] The algorithm searches in the said Atom table for the stationary phase, brand and grafting type fields identical to those of the analytical column entered (box 13).

[0122] If it recognizes it, the algorithm calculates the proximities (Euclidean distance) of the selectivities of all the stationary phases that are in the Atom table. It classifies the responses in ascending order and only retains candidates whose proximity values ​​are less than 0.17.

[0123] The algorithm compares its results to the Atom table and retains only six candidates according to criteria based on the number of injections and their presence in the column table, of which at least three columns are in physical stock at the user's.

[0124] The following steps are those described for box 20.

[0125] If the user does not have a suitable column (box 23), they are notified that they must purchase a suitable column. Depending on the column purchased, the following steps are those described for box 20.

[0126] If no solution is found following normal phase screening, but there is a reverse phase solution (box 24), we enter a second series of situations which concerns a non-polar or moderately polar (NP / MP) sample.

[0127] According to a first possibility, an analytical chromatogram of the separation is available (box 25).

[0128] The implementation of the method according to boxes 26-30 is similar to that described above with reference to boxes 10-14 and is therefore not described again.

[0129] In the case where no reversed-phase solution is available (box 31), the system proposes the implementation of a separation study on different reverse-phase selectivities, from an HPLC or flash chromatography system (box 32).

[0130] An algorithm allows in two steps and from a set of columns whose thermodynamic parameters p and q are known to determine the column with the most interesting separation potential (box 33). The first step determines the mobile phase conditions for which the retention factor of the last of the compounds eluted from the mixture to be purified is of the order of 10 and allows to recalculate the isoeluance conditions so that each of the other columns achieves this same objective. The second step is that of the injection of the sample on each of the columns at isoeluance, then by comparison of the chromatographic profiles, the most relevant column is chosen.

[0131] According to a first possibility, an analytical chromatogram of the separation is available (box 34).

[0132] The implementation of the method according to boxes 35-39 is similar to that described above with reference to boxes 10-14 and 26-30 and is therefore not described again.

[0133] In all cases, once the chosen column has been placed in the chromatography system, the processor starts the implementation of the chromatography according to the calculated operating mode (box 8).

[0134] Particularly advantageously, the preparative liquid chromatography process can be accelerated by means of the algorithm described below.

[0135] This acceleration possibility allows the user to satisfy, on the one hand, the need to complete a purification more quickly in order to be able to take care of another task that has become a priority and / or on the other hand to reduce the analysis time by estimating that the resolution of the critical pair will not be significantly impacted by the increase in throughput.

[0136] To this end, an algorithm evaluates the maximum admissible flow rate for the current purification and the pressure factor (taken into account in particular the limiting pressure of the chromatography system concerned, that of the column and possibly that of the Dry-load) according to the difficulty of the separation Δk for the TLC analytical data and Rs for the HPLC analytical data (box 40). An algorithm automatically adjusts the gradient method in real time to the new condition. The user can choose to activate the acceleration at any time during the purification. This operation is reversible and is carried out safely, without any loss of product.

Claims

1. Method for controlling preparative liquid chromatography, comprising the following steps, at least a part of said steps being implemented by a computer comprising a processor and a display screen coupled to said processor: (a) selecting, by means of an interface, an analytical liquid chromatography method from among thin layer chromatography (TLC) and high performance liquid chromatography (HPLC), (b) inputting analytical liquid chromatography data obtained by the method selected at step (a) for a product to be purified, (c) accessing a table of separating tools available to the user to implement said preparative liquid chromatography, said tools being columns, cartridges, or other chromatography means comprising a stationary phase, (d) from said analytical liquid chromatography data and table of available separating tools, selecting an optimal separating tool from said table and computing preparative liquid chromatography operating conditions for said selected separating tool.

2. Method according to claim 1, further comprising a step of entering the amount of product to be purified.

3. Method according to one of claims 1 or 2, further comprising a step of selecting the mode of introduction of the product to be purified into the selected separating tool, said mode of introduction being selected from introduction in solution or solid deposition.

4. Method according to one of claims 1 to 3, comprising a step of validation of the computation by a user and, after said validation, a step of launching the preparative liquid chromatography on the selected separating tool.

5. Method according to one of claims 1 to 4, comprising the triggering by a user of an acceleration of the preparative liquid chromatography, step (d) comprising the computation of optimised operating conditions while taking account of the maximum flow rate and the limit pressure of the preparative liquid chromatography system in which the selected separating tool is implemented.

6. Method according to one of claims 1 to 5, wherein the method selected at step (a) is thin layer chromatography (TLC) and step (b) comprises the input of analytical data obtained on a thin layer chromatography plate.

7. Method according to claim 6, wherein the selection and the computation implemented at step (d) take into account four different cases according to the retention factor (Rf) values of the plate: 0.01 < Rf < 0.07 , 0.08 < Rf < 0.4 , 0.4 < Rf < 0.9 , 0.9 < Rf < 1.

08. Method according to claim 7, comprising an adaptation of the shape of the elution gradient to each value by a volume difference factor (ΔCV) while taking account of the fact that a same factor (ΔCV) corresponds to differences in retention factor (ΔRf) different in each of said four zones.

9. Method according to one of claims 6 to 8, wherein the selection and the computation implemented at step (d) take into account from one to all of the compounds of interest spread out over the whole of the plate.

10. Method according to one of claims 6 to 9, wherein step (d) takes into account the difference (Δε°) between the eluant forces of the solvents used in the preparative liquid chromatography and the non-linear variation of the eluant force of a mixture of solvents as a function of the composition of said mixture.

11. Method according to one of claims 6 to 10, comprising the acquisition of an image of a thin layer chromatography plate and the automatic detection, from said image, of the compounds separated by a portable apparatus of the user, the input of analytical data of step (b) comprising an importation of data from said portable apparatus.

12. Method according to one of claims 1 to 5, wherein the method selected at step (a) is high performance liquid chromatography, the selection further comprising the choice of a chromatography mode from among: - normal phase liquid chromatography (NPLC), - reverse phase liquid chromatography (RPLC), - hydrophilic interaction chromatography (HILIC), - hydrophobic interaction chromatography (HIC).

13. Method according to claim 12, wherein the analytical chromatography data entered at step (b) are analytical liquid chromatography data obtained on a reference separating tool.

14. Method according to claim 13, wherein the table of available separating tools comprises a separating tool identical to the reference separating tool, and step (c) comprises the selection of said tool and the computation of the preparative liquid chromatography operating conditions for said separating tool.

15. Method according to claim 13, wherein the table of available separating tools does not include a separating tool identical to the reference separating tool, and step (d) comprises the selection of a separating tool having a stationary phase different from that of the reference tool by comparison of similarities between the stationary phase of the reference tool with the stationary phase of the available separating tools, or the selection of a separating tool comprising the same stationary phase as the reference tool and the computation of the preparative liquid chromatography operating conditions in multi-injection mode while minimising the number of injections.

16. Method according to one of claims 6 to 15, comprising a step of computation of separation by thin layer chromatography or by high performance liquid chromatography at iso-elution to determine other analytical conditions.

17. Method according to one of claims 6 to 16, wherein the selection and the computation of step (d) take into account the eluant force specific to each strong solvent.

18. Method according to one of claims 1 to 4, comprising the implementation of a separation test on a plurality of separating tools and the selection of the separating tool having the best separation potential from among said tools.

19. Method according to claim 18, wherein the table of available separating tools comprises a separating tool identical to the tool having the best separation potential, and step (d) comprises the selection of said tool and the computation of the preparative liquid chromatography operating conditions for said separating tool.

20. Method according to claim 18, wherein the table of available separating tools does not include a separating tool identical to the tool having the best separation potential, and step (d) comprises the selection of a separating tool comprising the same stationary phase as the tool having the best separation potential and the computation of the preparative liquid chromatography operating conditions in multi-injection mode while minimising the number of injections.

21. Method according to one of claims 1 to 20, comprising a sequencing of at least two series of steps (a) to (d), with two different analytical liquid chromatography methods and / or modes.

22. Method according to claim 21, wherein said sequencing is carried out automatically when a first series of steps (a) to (d) has not made it possible to identify a preparative liquid chromatography solution.

23. Preparative liquid chromatography system, comprising: - a computer comprising a processor, a user interface and a display screen, the processor being configured to implement the steps of the method according to one of claims 1 to 22, - a preparative liquid chromatography system controlled by said computer, comprising a mobile phase tank, a pump, an injector, a separating tool, a detector, a data recorder and a fraction collector.

24. Computer programme product comprising programme code instructions recorded on a support readable by a computer, characterised in that it comprises instructions for the implementation of the method according to one of claims 1 to 22 for a preparative liquid chromatography system according to claim 23.