Method and apparatus

Abstract

An aspect of the disclosure provides a method of biophysical characterisation of a lipid nanoparticle analyte, said method comprising the steps of: i) obtaining wave data for a sample based on physica

Description

Field of Invention The present invention relates to methods and apparatus, and more particularly to methods and apparatus for biophysical characterisation of lipid nanoparticle analytes. Background The use of lipid nanoparticles (LNPs) for the delivery of therapeutic compounds, in particular nucleic acids such as mRNA, has been intensifying since the successful rollout of LNP-based Covid-19 vaccines in 2021. Lipid nanoparticles are nanoparticles that arise from an amalgamation of lipid components that settle into an energetically favourable three-dimensional complex particle of its constituents. For instance, lipid nanoparticles may be colloidal structures comprising an amalgamation of lipids that exist in polymorphic liquid crystalline phases and comprise an amorphous lipid matrix, that can be used to carry a therapeutic payload. The properties of lipid nanoparticle formulations, precursor lipid nanoparticle formulations and constituent biomolecules thereof, and the manner in which they will perform in certain circumstances of use, may be difficult to determine. LNP properties such as their size, structure and. electrostatic properties are often fundamental to the performance and efficacy of an LNP, for instance for use as a therapeutic delivery mechanism. A major area of research is so-called "developability assessment", which relates to the evaluation of a potential therapeutic candidate's properties. Often the objective of such evaluation is to avoid late-stage failures in clinical research and to improve; the manufacturability and safety profile of drugs and. therapeutic compositions, such as those comprising lipid nanoparticles. The surface of a material has a thermodynamic potential that is independent of its volume. The physical and chemical properties of a surface are derived from its thermodynamic potential. For example, the response of the surface to a mechanical perturbation is given by properties such as surface tension and lateral compressibility. Similarly, the response of the surface to an electromagnetic perturbation is given by properties such as surface dipole moment. As a result of these perturbations, different types of surface waves may be generated on a surface e.g. a surface of a fluid (e.g. a liquid) forming an interface with another fluid (e.g. air). Some example; types of surface waves are: Rayleigh waves; Gravity waves; Capillary waves; Lucassen waves. The physics of these waves have been described in Nonlinear fractional waves at elastic interfaces Julian Kappler , Shamit Shrivastava, Matthias F. Schneider , and Roland R. Netz Phys. Rev. Fluids 2, 114804 - Published 20 November 2017. These waves may be hydrodynamical1y coupled. Rayleigh 'waves are characterised by elliptical motion of a notional fluid particle in a plane which is perpendicular to the surface at equilibrium and parallel to the direction of propagation of the wave. Gravity waves are characterised by a displacement from equilibrium of a notional fluid particle at the surface wherein the displacement of the notional particle is characterised by having a restoring force of gravity or buoyancy. Capillary waves are characterised by a displacement from equilibrium of a notional fluid particle wherein the displacement of the notional fluid particle is in a direction transverse to the surface at equilibrium and transverse to the direction of propagation of the wave and have a restoring force of surface tension. Lucassen waves are characterised by a displacement from equilibrium of a notional fluid particle at a surface of a wave-medium by oscillation in a direction parallel to that surface at equilibrium and parallel to the direction of propagation of the wave. In Lucassen waves this notional particle is subject to a restoring force resulting from the surface elastic modulus of the surface of the wave-medium. Put another way Lucassen waves are compression-rarefaction waves which occur in the plane of a boundary (an interface) between a wave-medium and. an adjacent medium such as air. Lucassen waves have been observed in lipid monolayers and in other types of liquid systems. Shamit Shrivastava, Matthias F. Schneider Opto-Mechanical Coupling in Interfaces under Static and Propagative Conditions and Its Biological Implications describes how a wave can be generated in a lipid monolayer mechanically with a dipper and how parameters of the generated wave, such as the intensity of fluorescent particles therein and the lateral pressure of the surface wave, can be measured, for example using a photo detector and a Wilhelmy balance respectively. Shrivastava S, Schneider MF. 2014 Evidence for two-dimensional solitary sound waves in a lipid controlled interface and its implications for biological signalling. J. R. Soc. Interface 11: 20140098 describes a method in which Lucassen waves can be generated in a lipid monolayer and how parameters of said waves may be measured (e.g. fluorescence energy transfer (FRET) measurements; a piezo cantilever) . The document also describes how the state of a lipid monolayer may be characterised by a variety of thin film parameters (e.g. surface density of lipid molecules, temperature, pH, lipid-type, ion or protein adsorption, solvent incorporation, etc.) and also now the state of the lipid monolayer can affect parameters of waves which propagate in the lipid monolayer. Bernhard Fichtl, Shamit Shrivastava &Matthias F. Schneider, Protons at the speed of sound: Predicting specific biological signaling from physics Nature Scientific Reports describes how Lucassen waves can be generated in a lipid interface in response to a change in pH of the system and that the speed of these waves can be controlled by the compressibility of the interface. The document describes how parameters of these waves depend on the degree of change in pH. The document also describes how mechanical and electrical changes at the lipid interface can be measured (e.g. using a Kelvin probe). Lucassen waves may be described as interfacial compression waves and may be considered two-dimensional sound waves (sound waves confined to a surface which forms a boundary between two phases e.g. a fluidair boundary). In a manner analogous to sound waves, shock waves may exist in Lucassen wave systems (e.g. two-dimensiona1 shock waves). Lucassen shock waves may be characterised in the same way as Lucassen waves with the additional constraint that the waves are characterised by changes in the wave medium which are nonlinear and / or discontinuous . S. Shrivastava, Shock and detonation waves at an interface and the collision of action potentials, Progress in Biophysics and Molecular Biology, describes how Lucassen shock waves may propagate through a lipid interface. WO2019234437A1 describes how a lipid interface may be used to transmit and receive signals. The document describes a signal processing device comprising: a first medium; a second medium; a lipid interface arranged between the first medium and the second medium, wherein the lipid interface comprises a plurality of lipid molecules; an input transducer arranged to apply an input signal to the lipid interface, wherein the input signal is arranged to generate a mechanical pulse in the lipid interface; and an output transducer arranged to receive an output signal by detecting a mechanical response in the lipid interface from the mechanical pulse generated in the lipid interface by the input transducer; wherein the lipid interface is arranged to propagate the mechanical pulse from the input transducer via the lipid interface to the output transducer. Summary Aspects of the invention are set out in the independent claims and optional features are set out in the dependent claims. Aspects of the disclosure may be provided in conjunction with each other and features a' ;ts . Brief Description of Drawings Embodiments of the disclosure will now be described in detail with reference to the accompanying drawings, in which: Figure 1 shows a flow chart indicating the steps of a method for biophysical characterisation of a lipid nanoparticle analyte; Figure 2 shows a diagram of an apparatus which may be used for performing the measurements of the physical wave in the liquid as described in embodiments of the present disclosure and comprising the hardware components employed in methods such as those described with reference to Figure 1 and / or Figure 2; Figure 3 shows a flow chart indicating the steps of a method of identifying a lipid nanoparticle analyte with comparable biophysical properties to a target lipid nanoparticle analyte; Figure 4 shows a flow chart indicating the steps of a method of biophysical characterisation of a biomolecule or composition thereof, wherein wave data is obtained based on physical measurements of a series of waves; Figure 5 shows a plot of displacement of the liquid at a location (Y-axis) against time (X-axis) for two different samples - the first sample being of one glyceride oil, the second sample being the same glyceride oil adulterated with 0.1% of a different glyceride oil; Figure 6 shows a visual representation of an array of 'wave data in which the row’s of the array each correspond to a 'waveform obtained by measuring a physical wave at a location in the liquid as a function of time; and Figure 7 shows a simplified visual representation of the identification of separable contributions of the variance in a data array such as that illustrated in Figure 6; Figure 8A shows the results of principle component analysis (PCA) from wave data obtained for a selection of constituent biomolecules; Figure 8B shows mean isotherms of surface pressure data generated by deposition of constituent biomolecules; Figures 9A, 9B, 9C and 9D show the results of principle component analysis (PCA) from wave data obtained for a selection of precursor lipid nanoparticle formulations . In the drawings, like reference numerals are used to indicate like elements. Specific Description Figure 1 illustrates a method of biophysical characterisation of a lipid nanoparticle analyte. The characterisation is based on wave data, obtained from physical measurement of a wave (or waves) in a liquid generated by contact between the liquid and a droplet of that 1ipid nanoparticle analyte. Initially, prior to beginning the method illustrated in Figure 1, a sample is prepared comprising the lipid nanoparticle analyte, as described in more detail below. In brief, the lipid nanoparticle analyte typically comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation. A droplet of the prepared sample is then brought into contact with the liquid. This may comprise dropping the droplet from a dispenser so that it drops onto and impacts the liquid or the droplet may be lowered into a position in which it contacts the liquid, such as by moving a nozzle of the dispen g the drople Th' oplet may also be generated in a position such that contact with the liquid occurs when the droplet reaches a given size. It will thus be appreciated that contact may take place at negligible net linear The liquid may be provided in a trough, such as a Langmuir trough or any similar system.. The characteristics of the liquid will be described in detail below, but typically it comprises a liquid solution (such as an aqueous solution) and / or liquid suspension. Suspensions including colloidal suspensions may be used. A thin film may be provided at the surface of the liquid. Contact between the droplet and the liquid generates a physical wave in the liquid. This wave is measured at a particular location, which typically is displaced a short distance from the location at which contact takes place. These measurements provide a time: series of samples which describe disturbance of the liquid at that particular location. Typically, the measurements are performed optically or electrically but, however it is achieved, the measurement technique generally provides measurement of the physical wave which includes sensitivity to at least two or more of the following wave modes: Rayleigh waves, gravity waves, capillary waves and Lucassen waves. Typically, a sensitivity to Lucassen waves is desirable. These wave modes encode characteristics of the interaction between the sample and the liquid system which themselves depend on the biophysical properties of the biomolecule or composition in the sample. The wave data is stored in a data store, such as a digital memory, for retrieval by a processor. Typically, the wave data associated with the droplet comprises a waveform or set of waveforms which provide a measurement of the physical wave in the liquid having sufficient degrees of freedom to encode the relevant wave modes which make up that physical wave. A visual representation of two such waveforms is illustrated in Figure The above procedure may be repeated for a plurality of droplets, so for each droplet, that waveform or set of waveforms is stored into the data store. The condition of the liquid for each droplet may be identical to all preceding droplets. For example, the composition of the liquid at the location where contact occurs may be controlled so that it remains substantially equal for all droplets. One way to achieve this is to provide flow in the liquid, but other methods may be used. It is also possible however to allow the lipid nanoparticle analyte to accumulate in the liquid, so that its concentration at the location where contact occurs increases as droplets are applied to the liquid. In any event, the end result of this procedure is a set of tuples of wave data. Each tuple corresponding to the waveform(s; obtained by measuring the physical wave caused in the liquid by a corresponding one of the droplets. This may be stored as an array of data in which the rows or columns correspond to such tuples. A visual representation of such an array is provided in Figure 6. This array of wave data is then operated on by a data processor, which may be configured to identify separable contributions to the waveforms described in that data. This may be done by the processor operating on the data to perform a dimensionality reduction of the data. For example, the processor may be configured to apply a principal components analysis (PCA) to the wave data. It v / ill be appreciated in the context of the present disclosure that such analysis generates a number of coefficients and a number of vectors. For example, the coefficients, p, may be the eigenvalues and the vectors, P, the eigenvectors of the covariance matrix of the wave data. Other dimensionality reduction techniques may be used to identify one or more; separable contributions to the; variance in the wave data. A visual representation of dimensionality reduction of wave data is provided in Figure 7. Next, association data is obtained from a data store, such as digital memory. The; association data may describe a predetermined relationship between characteristics of the wave data and one or more biophysical characteristics of the sample. For example, it may provide a mapping from the separable contributions to particular values of the biophysical characteristics. The method may comprise the data processor obtaining, e.g. from input, information defining the class of lipid nanoparticle analyte and selecting the relevant model from the association data according to the class of lipid nanoparticle analyte. For instance, where the lipid nanoparticle analyte comprises a lipid nanoparticle formulation, the association data may store a set of models for any one or more of the following classes of lipid nanoparticle analytes: ® lipid nanoparticle formulations, such as: o liposomes, such as conventional liposomes, sterically-stabilized liposomes, and ligand-targeted liposomes; o solid lipid nanoparticles; o nanostructured lipid carriers; o lipid polymer hybrid nanoparticles; • constituent biomolecules of a lipid nanoparticle, such as: o structural lipids, such as cholesterol and derivatives thereof; o cationic and / or ionisable lipids, such as monovalent or multivalent aliphatic lipids; o phospholipids, such as ph o spha t i dy1ch o1i nes, p h o s p n a t i d y 1 e t h a n o 1 a mi ne s, p h o s p h a t i dy 1 s e r i n e s, p h o s p n a t i d y 1 i n o s i t o 1 s, p h o s p h a t i d y 1 g 1 y ce r ol s , an d phosphatidic acids; o PEGylated lipids, such as PEG-modified p h o s pha t idy1et ha no1ami ne s , PEG-modified phospha t i di c acids, PEG-modified ceramides, PEG-modif ied dialkylamines, PEG-modified diacylglycerols, and PEG-modified dialkylglycerols; o surfactants, such as non-ionic surfactants, anionic surfactants, and amphoteric surfactants; o tri-, di-, and mono-glycerides; o fatty acids; and o waxes, and • precursor lipid nanoparticle formulations,, such as mixtures of two or more of the constituent biomolecules as defined above . It will be appreciated in the context of the present disclosure that these are merely examples of the classes of lipid nanoparticle analytes to which the present disclosure may be applied. The methods of the present disclosure are able to provide biophysical characterisations for any nanoparticulate lipid composition. The data processor may also be configured to select within the set of models stored in the association data according to the biophysical parameter (or parameters) which is / are to be determined for that class Next, the association data for the relevant class of lipid nanoparticle analyte and the relevant biophysical characteristic (s) is used, with the vrave data, to determine data indicating those biophysical characteristic (s) . This may comprise performing operations, defined in the association data, on data indicating the separable c o n t r i b u t i o n s. An apparatus to implement this method is depicted in Figure 2. In this example, the source of the wave data in the apparatus illustrated in Figure 2 is an analytical instrument 1 configured and arranged to perform physical measurements of the wave in the liquid which is generated by contact between the liquid and a droplet of the sample. As illustrated, the instrument 1 is configured to generate the wave data as described below and to provide that wave data to the data processor. The data processor comprises an input connection for obtaining the wave data from a data source such as the instrument 1. The data processor is also connected to a data store, such as a digital memory, storing' the a s s o c i a t i o n data. The association data may be based on prior analyses of other lipid nanoparticle analytes of the same class as those which are present in the sample. As illustrated, the association data may comprise a plurality of sets of models - each set of models relating to a certain class of lipid nanoparticle analytes. Each such model comprises digital data defining data operations to be performed by the processor to provide a mapping between the wave data and one or more biophysical characteristics . A variety of such models may be used. As a first example, the association data may comprise a plurality of coefficients defining a linear model of one or more biophysical characteristics as a function of selected separable characteristics of the wave data. The data processor may be configured to determine the biophysical characteristic by scaling each of the selected contributions, p, according to a corresponding one of the coefficients in the association data and then making a linear sum of the scaled contributions. As a second example, the model may be non-linear and may define the one or more biophysical characteristics as a non-linear function of the selected contributions, p, and a data processor may apply the non-linear function to those separable contributions, p, to determine the biophysical characteristic. As a third example, the association data may define an adaptive model, for example it may comprise a machine learning element in the form of a network of nodes and / or weights and the data processor may be configured to apply that machine learning element to the wave data to determine the biophysical characteristic of the lipid nanoparticle analyte in the sample. The nodes in that network may each comprise the linear and / or non-linear operations as described above and may also comprise dimensionality reduction processes to identify separable components for use in such operations. For example, the association data and the data processor together may be configured to implement a machine learning element, such as a neural network or other machine learning system. The weights of such a machine learning element may be retrieved from a reference database, and may have been learned from prior wave data, which may have been obtained using reference data, such as may have been obtained from wave data for known comparator substances / analytes, such as known LNP analytes. An appropriate reference database may comprise wave data and / or weights and / or a machine learning element comprising such weights generated known comparator substances / analytes. The reference database may comprise metadata corresponding to the comparator analyte, which may include data identifying the comparator and indications of biophysical properties or attributes for the comparator analyte. The known comparator may be a target lipid nanoparticle analyte or may be selected as having one or more biophysical characteristics in common with such a target. The association data may comprise a set of such models for each class of constituent biomolecule, and for each class of lipid nanoparticle formulations and precursor formulations comprising constituent biomolecules of that class. For example, the association data may comprise; a set of models for structural lipids, a set for cationic and / or ionisable lipids, a set for phospholipids, a set for PEGylated lipids, and so on. The association data may comprise a set of models for precursor lipid nanoparticle formulations comprising various combinations of these constituent biomolecules, as well as a set of models for nano structured lipid carriers, a set. for liposomes, and so on. Each model in each set may relate one or more predetermined biophysical characteristics of lipid nanoparticle analytes of that class to characteristics of the wave data for lipid nanoparticle analytes biomolecules of that class, such as selected separable components of such wave data. The data processor is configured to obtain, e.g. from input or hardcoded data, information defining the class of lipid nanoparticle analyte biomolecule. It is further configured to select the relevant model from, the association data according to the class of lipid nanoparticle analyte. The data processor may also be configured to select within the set according to the biophysical parameter (or parameters) which is / are to be determined for that class. As a first example, where the sample comprises a lipid nanoparticle analyte which is a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, the data processor may be configured to select the association data related to lipid nanoparticle formulation classification and to select within that association data, the model(s) associated with a metric of developability, such as manufacturability; in vivo performance such as biodistribution, circulation time, reduction in mortality or morbidity, protein expression level; in vitro performance such as transfection efficacy; encapsulation efficiency; polydispersity index; lipid nanoparticle size; stability; or safety profile. As a second example, where the sample comprises a lipid nanoparticle analyte which is a constituent biomolecule of a lipid nanoparticle composition, such as a PEGylated lipid or an ionisable and / or cationic lipid, the data processor may be configured to select the association data related to quantitative properties selected from viscosity, phase state in a particular environment, stability in a particular environment, hydrophobicity, binding affinity to a particular target, solubility in a particular solvent, surface charge, and specific gravity. processor is configured to perform data operations defined by the selected association data for the relevant class of lipid nanoparticle analyte on the wave data to determine data indicating the biophysical characteristic. .As described above, dimensionality techniques can be used to identify separable contributions of the wave data and then the separable contributions can be used to draw inferences about the biophysical characteristics of the lipid nanoparticle analyte. Examples of dimensionality reduction method comprises a blind signal separation, BSS, methods such as: principal component analysis; singular value decomposition; independent component analysis; and non-negative matrix factorization. Other examples of dimensionality reduction methods comprise spectral analysis and multi-resolution analysis methods, such as time-frequency analysis (including analysis based on a short-time fourier transform and wavelet analysis). However, it is not necessary to use a separate dimensionality reduction step or indeed to perform dimensionality reduction at all. A machine learning element, such as a neural network or other machine learning system, may comprise; an embedding to reduce its dimensionality as part of its integrated data processing functions rather than as a separate preceding step. In some embodiments a machine learning element may be trained to attribute one or more biophysical characteristics of lipid nanoparticle analytes to wave data. In these embodiments, there is no need to determine the separable contributions by dimensionality reduction or otherwise. The association data may define a relation between the wave data and the biophysical characteristic. Machine learning is just one way to do this -- the association data may define other mappings between vrave data and biophysical characteristics so as to act on the wave data to provide a biophysical characterisation of the 1ip1d nanoparticle analyte. It will be appreciated in the context of the foregoing disclosure that the data processor may be configured to act on data obtained directly from an instrument 1 such as that illustrated in Figure 2. For example, the methods of the present disclosure may comprise performing the physical measurements of the wave, such as by preparing the sample for analysis, providing the contact between the droplet of the sample and the liquid system to generate the wave, and then also performing the physical measurements of the wave. However, the physical measurement may have been performed separately by another instrument, and the data processor may be configured to act on data obtained from other sources such as retrieval from memory or received via a netvrork message . The physical wave may comprise a surface; wave; comprising modes encoding characteristics of the interaction between the sample and. the liquid. Examples of such wave modes include a Lucas sen wave mode. Whatever or measurement method is used, the physical measurement may comprise a sens including a Lucassen wave mode. such inodes, optiona 1ly The instrument 1 is one example of an apparatus for characterising an interaction between a droplet of a sample and a liquid system. The apparatus characterises a wave 5 in the liquid system, based on physical measurement of the wave. This may be done by optical methods such as by measuring the polarisation of a light beam 7 reflected by the wave in particular it may use a ratio between the s-polarisation component and the p-polarisation component of the reflected light beam 7 to sense variations in refractive index (and hence density) of the liquid in the presence of the wave. A time series of such data at a particular location in the liquid can be used to characterise waves. The apparatus 1 shown in Figure 2 comprises, a droplet provider 9 configured to contact the liquid with a droplet of a sample, light beam optics 11, a light collector 13, a detector 15, and a wave measurement module 17. As illustrated in Figure 2, the apparatus 1 also comprises a reservoir 23 holding a volume of liquid 21 (i.e. the reservoir medium). The reservoir 23 may be provided by a trough, such as a Langmuir trough. Also shown in Figure 2 mechanical fixtures 19 may be provided for holding the apparatus in position with respect to the reservoir 23, but it 'will be appreciated that these fixtures 19 are not essential and may be made and sold separately from the apparatus 1 itself. The wave measurement module 17 is connected to the droplet provider 9 and to the detector 15 for the communication of control signals and data, it may also be connected to the light beam optics 11. Examples of types of liquid 'which may provide the volume of liquid 21 (i.e. the reservoir medium) comprise; aqueous solutions (e.g. deionized and purified water) and suspensions. The light beam optics 11 comprise a source of polarised light arranged to illuminate an area of the liquid surface 3 with a beam 7 having a selected angle of incidence a. The light beam 7 may also be coherent. Examples of suitable light sources include lasers and. the light beam optics may comprise a polariser. The light collector 13 is arranged to receive the beam of light 7 after reflection by the area of the thin film and to provide the reflected beam, of light to the detector. The light collector 13 is positioned so that the optical axis of the light collector 13 is directed to the area of the liquid surface 3 at the angle of specular reflection, a, of the light beam incident on the liquid surface from the light beam optics. The detector 15 is configured to sense parameters of the light received from the light collector 13 and to provide signals to the wave measurement module 17 including those parameters. Typically, those parameters comprise parameters capable of characterising the wave and may comprise a sensitivity to a variety of wave modes such as Lucassen waves and other types of wave. For example, the parameters may indicate the polarisation of the light beam 7 after interaction with the wave. For example, the parameters may comprise a measure of the intensity of one or more polarisation components of the received light, such as the intensity of (a) a first component of the second polarisation and / or (b; a second component of the second polarisation. The second component may be orthogonal to the first component. The first component may be the s-component and the second component may be the p-component. The droplet provider 9 is positioned with respect to the liquid 5 so that it can provide contact between the liquid and a droplet of the sample. For example, the droplet provider 9 may comprise a source of the sample and may be configured to contact the surface of the liquid with a droplet of the sample; to generate a wave in the liquid. Such stimulus may create a wave 5 in the liquid exhibiting some or all of the above wave modes. The wave measurement module 17 is configured to control the droplet provid er 9 to apply the stimulus to the liquid, and to operate the detector 15 to collect a time series of samples of the light received at the detector 15. These samples may comprise samples of the intensity of the one or more polarisation components mentioned above. Typically., the sample rate is at least 1MHz, for example 10MHz. The wave measurement module may also be configured to apply a low pass filter to the time series before down-sampling the data to 20kHz or thereabouts. Typically, the sample rate of the down-sampled time series is selected based on the size of the illuminated area and the expected speed of the wave in the liquid surface. For example, the expected speed may be approximately 1 ms-1 and the illuminated area may have a diameter of ~5mm, in which case the upper limit on the frequency of surface waves that can be meaningfully sampled will be 10kHz. The sample rate of the down-sampled time series may be selected to ensure that the measurement remains well within this available bandwidth. The wave measurement module 17 may be configured to control the timing of these samples based on the operation of the droplet provider 9, for example so that the surface wave 5 in the area of the liquid surface illuminated by the beam 7 can be sampled at a selected time after the contact with the droplet and for a selected duration. The time and / or duration typically are selected based on the distance from the part of the liquid to which the droplet is applied to the illuminated area. The wave measurement module may be further configured to provide a particular sampling scheme for a particular measurement type. The wave measurement module may be configured to implement a first sampling scheme to perform a first measurement type and to implement a second, different, sampling scheme to perform a second, different, measurement type. For example, where a thin film is present at the surface of the reservoir, to measure viscosity or hydrophobicity in a lipid thin film, or to measure binding in a protein thin film, the wave measurement module may use a long sampling duration (total time for which samples are collected). Recording of a single stimulus typically has a time resolution of microsecond and duration of seconds. This can be sufficient for measurement of properties of molecule that interact strongly with the film and / or are fast, for example electrostatic interaction or hydrogen bonding. In some embodiments multiple such stimulations with a repetition rate of few seconds observed over a course of minutes to hours could provide improved measurement of properties of molecules that interact weakly and / or slowly with the film, for example binding or reaction kinetics. The wave measurement module may be configured to sample data in a selected time interval following the application of a stimulus to the liquid surface of the reservoir, and to repeat the same sampling in that same time interval after subsequent stimuli to provide repeated measurements. Such measurements may be of relatively short duration. In operation, the wave measurement module 17 operates the droplet provider 9 to provide contact between the liquid and a droplet of the sample. This triggers a wave 5 in the liquid. The wave travels outwardly, across the liquid from the location at which the contact takes place. The light beam optics 11 illuminate an area of the liquid through which the wave travels, and the wave measurement module 17 operates the detector 15 to take a series of samples which represent the light beam 7 after its interaction with the area. These may be provided by the light collector 13 to the detector 15. Accordingly, the disturbance of the liquid by the wave at the location can be recorded as a function of time in a series of samples of data (a time series; . Each sample in that time series may comprise polarisation data, which may be in the form of the intensity of the s-polarisation component and the intensity of the p-polarisation component of the reflected light. The vrave measurement module 17 may be configured to determine an indication of the polarisation angle of the reflected beam 7, such as a ratio of the intensity of the s-component to the intensity of the p-component for each sample. The wave measurement module may derive features of the Lucassen wave from this time series. Examples of features of the Lucassen wave include its amplitude, frequency content, phase velocity, group velocity, phase and so forth. The wave measurement module 17 may then use these features of the Lucassen wave to provide information about the stimulus or about an optional thin film at the reservoir surface, as described below. Other methods of wave measurement may be used. This is just one example. Imaging methods and non-optical methods may be used including x-ray methods, electrical transducers and mechanical transducers. The apparatus may be controlled to provide a series of such measurements of a series of such waves, each wave being generated by contact between the liquid and a droplet of the test substance. This may generate wave data comprising a plurality of individual waveforms, each corresponding to the physical measurement of one wave, generated in the liquid by contact with the droplet of sample. It will be appreciated in the context of the present disclosure that the change in polarisation caused by reflection at an interface is related to the refractive index differences at the interface. The inventors in the present case have appreciated that refractive index may be related to the surface characteristics of a liquid. Accordingly, the wave measurement module can derive, from the time series of samples, such as the s-p ratio, information about variations in those characteristics as a function of time. This may enable complex wave m.odes with many degrees of freedom to be characterised. The reservoir surface may carry a thin film, and the characteristics of the interface which are measured to generate the wave data may be associated with variations in density and / or position of that thin film, as well as the orientation of molecules within the film. As will be appreciated, following its deposition, the lipid nanoparticle analyte itself may populate a pre-existing thin film, or a thin film may be generated from the lipid nanoparticle analyte itself. Where a thin film is present typically, the thin film comprises a type of liquid which is different from that of the volume of liquid 21. The liquid thin film and the volume of liquid 21 may therefore have an interface between them such as a liquid-liquid interface. The thin film may have viscoelastic properties. These and other types of thin films may exhibit a variety of surface wave modes in response to stimulus. Examples of such wave modes comprise Rayleigh waves, gravity waves, capillary waves and Lucassen waves. Examples of types of liquid which provide the thin film include proteins and lipids and other types of liquid. It will be appreciated in the context of the present disclosure that such materials may also be held (e.g., dispersed in suspension or otherwise) in the volume of liquid and dynamic equilibrium may exist between the thin film, and the material held in the volume of liquid 21. The above apparatus may be used in other methods. One such method is depicted in the flow chart illustrated in Figure 3. Figure 3 illustrates a method of identifying a lipid nanoparticle analyte within a class of lipid nanoparticle analytes with comparable biophysical properties to a target lipid nanoparticle analyte of the same class of lipid nanoparticle analyte. As described above with reference to Figure 1, a sample comprising a lipid nanoparticle analyte within a class of lipid nanoparticle analytes may be prepared and contact may be provided between a droplet of such a sample and a liquid. The physical wave generated in the liquid by that contact may be measured to obtain physical wave data as also described above. However, that wave data which is obtained comprises modes encoding characteristics of the interaction between the sample and the liquid system. A data processor, such as that described above with reference to Figure 2 then provides a comparison of the wave data obtained from interaction of the sample with the surface of the liquid system with wave data associated with the target lipid nanoparticle analyte. This comparison may be based on (a) a biophysical characterisation of the lipid nanoparticle analyte based on the wave data; and (b) a corresponding characterisation of the target lipid nanoparticle analyte. Any of the biophysical characterisations described above may be used. For instance, the corresponding characterisation of the target lipid nanoparticle analyte may be derived from wave data obtained from interaction of the target lipid nanoparticle analyte with the surface of the liquid system, or it may be derived from wave data generated by an LLM model. The above steps can then be repeated for each of a number of different lipid nanoparticle analytes of the same class until a lipid nanoparticle analyte is identified having values of selected biophysical characteristics, indicative of properties of the lipid nanoparticle analyte, which match those of the target lipid nanoparticle analyte. The nature of this matching process may depend on the class of lipid nanoparticle analyte in question. For example, developability of therapeutic candidates may be assessed in this way. As described herein, in one aspect, there is provided a method of identifying a lipid nanoparticle analyte within a class of lipid nanoparticle analytes with comparable biophysical properties to a target lipid nanoparticle analyte of the same class of lipid nanoparticle analyte. The method involves a comparison of the wave data obtained from interaction of a sample of the lipid nanoparticle analyte, with the surface of a liquid system with wave data associated with the target lipid nanoparticle analyte. In another aspect, a method is provided in which a biophysical characterisation of a lipid nanoparticle analyte is made based on wave data derived from the interaction of a lipid nanoparticle analyte sample and a liquid system. The results of the biophysical characterisation may be incorporated as reference data within a library, against which the wave data of other lipid nanoparticle analytes may be compared. In order to facilitate comparisons between wave data obtained for a target lipid nanoparticle analyte, or reference lipid nanoparticle analyte held in a library, the lipid nanoparticle analyte sample which is tested is selected such that a liquid carrier present in the sample: is substantially the same as that which is employed in acquiring the corresponding wave data of the target or reference lipid nanoparticle. Similarly, consistency in the liquid medium of the reservoir of the liquid system, as well as height of the droplet provider and volumetric flow rate thereof discussed herein, also facilitate a comparison. Nevertheless, this is not essential since comparisons may still be made ’with contributions from any variation in the method of wave data acquisition being accounted for computationally (e.g. by multidimensional calibration). Figure 4 illustrates a further method, of biophysical characterisation of a biomolecule, or a composition thereof. As described above with reference to Figure 1, a sample comprising a biomolecule, or composition thereof, may be prepared and contact may be provided between a droplet of such a sample and a liquid. The physical wave generated in the liquid by that contact may be measured to obtain physical wave data as also described above. The wave data which is obtained comprises modes encoding characteristics of the interaction between the sample and the liquid system. However, in this method, a series of waves are measured, with each wave in the series being generated by providing contact between one sample droplet of a series of sample droplets and the liquid system. As such, the wave data used to provide a biophysical characterisation of the biomolecule, or composition thereof, comprises wave data from a plurality of waves in the series --- for instance, the wave data used may comprise substantially all of the wave data generated by the series of droplets. As part of this method, when contact is provided, between the sample droplet and the liquid system, the sample deposited on the liquid, system is allowed to accumulate on the surface of the liquid system. In this way, subsequent sample droplets are contacted with a liquid system that comprises incremental!y higher concentrations of sample, and. in particular higher concentrations of the biomolecule, or composition thereof, contained therein. This accumulation eventually leads to the formation of a thin film comprising the biomolecule, or composition thereof, on the surface (i.e. at the air-liquid interface) of the liquid system. The thin film need not extend across the entirety of the surface of the liquid, but may form a saturated layer across at least a portion of the liquid system. Once the thin film has formed, at least one of the sample droplets of the series of droplets is contacted with the thin film, and the wave data generated thereby is preferably used as part of that used to provide the biophysical characterisation. The biophysical characterisation may be provided in the same manner as described herein . The method described in Figure 4 allows for biophysical characterisation of biomolecules, or compositions thereof, based on wave data that is obtained in connection therewith. Biomolecules are well-known organic molecules, primarily composed of hydrogen and carbon, which are found in nature, typically in organisms and / or form part of biological processes. There is no particular limit on the nature of the biomolecules that can be used in the methods described herein, or compositions comprising them. Preferably, however, the biomolecule or composition thereof is a lipid nanoparticle analyte, which comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, as described herein. Thus, the sample preferably comprises the lipid nanoparticle analyte, wherein the lipid nanoparticle analyte comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation. In addition, where the sample comprises a lipid nanoparticle analyte, the sample preferably further comprises an organic solvent or solvent system, such that the lipid nanoparticle analyte is dissolved in the organic solvent or solvent system.. In embodiments comprising an organic solvent in the sample droplet, the solvent will be deposited onto the surface of the liquid system alongside the biomolecule or lipid nanoparticle analyte. Thus, the thin film generated by deposition of the sample may comprise the organic solvent. Preferably, the solvent is a volatile solvent, such as chloroform, that will readily evaporate from the surface of the liquid system once deposited, leaving only the biomolecule or lipid nanoparticle analyte to form the thin film. Accordingly, in this method the thin film may be generated solely from the biomolecule, in particular from the lipid nanoparticle analyte. As part of this method, the biophysical characterisation may be further based on wave data generated during the phase when the thin film was still being deposited, i.e. before the thin film is fully formed. This wave data, indicating the build-up of the analyte on the liquid system, may describe the surface pressure of the liquid system. Changes in intensity and polarisation of light that is reflected off the reservoir surface can also be indicative of the build-up of analyte. Thus, in one aspect, the present disclosure provides a method of biophysical characterisation of a biomolecule or composition thereof. The method comprises obtaining wave data for a sample based on physical measurements of a series of waves, each wave of the series being generated, in a liquid system, by providing contact between a series of droplets of the sample and the liquid system, wherein the waves comprise modes encoding characteristics of the interaction between the sample and the liquid system, and wherein the sample comprises the biomolecule, for analysis. The method further comprises providing a biophysical characterisation of the biomolecule, or composition thereof, based on the wave data. As part of this method, providing contact betwreen the series of droplets of the sample and the liquid system generates a thin film comprising the biomolecule, or composition thereof, on the liquid system, and at least one of the droplets of the series of droplets is contacted with the thin film. In one embodiment of this aspect, following formation of the thin film by deposition of a first sample, contact between the thin film and droplets of a second sample is provided, wherein the second sample comprises a different biomolecule. For instance, the second biomolecule may be an excipient, a therapeutic compound or a protein (as described below). The wave data generated by this contact can be obtained, and can be used to provide a biophysical characterisation of the interaction between the biomolecules of the first and second samples . Further discussion of particular lipid nanoparticle analytes useful in the methods described herein, together with discussion of sample preparation is provided below. Lipid nanoparticle formulations The methods described herein allow for biophysical characterisation of lipid nanoparticle analytes based on wave data that is obtained in connection therewith. The lipid nanoparticle analyte comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation. Preferably, the lipid nanoparticle analyte comprises a precursor lipid nanoparticle formulation. Preferably, the lipid nanoparticle formulations, precursor lipid nanoparticle formulations and constituent biomolecules thereof are not, and more preferably do not comprise, glyceride oils, such as plant or animal oils and fats. A glyceride oil refers to an oil or fat which comprises triglycerides as the major component thereof. A lipid nanoparticle, as used herein, refers to a spheroid nanoparticle which is composed predominantly of lipids, such as might be used as a therapeutic delivery system. Lipid nanoparticles typically have an average diameter in the range of from, about 10 to 1000 nm, with average diameters in the range of about 10 to 200 nm being most commonly used for therapeutic purposes. The diameter of a lipid nanoparticle can be measured using dynamic light scattering (DLS), as described in ISO 22412:2017. Examples of classes of lipid nanoparticles include liposomes, solid 1ipid nanoparti cles, nanostructured 1ipid carriers, and 1ipid po1y me r hybrid nanoparticles, each of which are intended to be encompassed by the term "lipid nanoparticle". Preferably, the class of lipid nanoparticle being characterised is a nanostructured lipid carrier. Accordingly, a lipid nanoparticle formulation refers to a formulation that comprises assembled lipid nanoparticles. For instance, the lipid nanoparticle formulation may be in the form of a microemulsion or a nanoemulsion, wherein the average diameter of a lipid nanoparticle therein is in the range of about 10 to 500 nm. Lipid nanoparticles may comprise a variety of lipids, depending on the particular class of lipid nanoparticle. For instance, a lipid nanoparticle may comprise any one or more of: a phospholipid; a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid; a non-phospho lipid surfactant; a tri--, di--, or mono---glyceride; a fatty acid; a wax; and combinations thereof. For instance, the lipid nanoparticle formulation may comprise at least one of: a structural lipid; a cationic and / or ionisable lipid; a phospholipid; and a PEGylated lipid. Preferably, the lipid nanoparticle composition comprises at least one or both of a cationic and / or ionisable lipid and a PEGylated lipid, and more preferably comprises a compound of each of the classes of: a structural lipid; a cationic and / or ionisable lipid; a phospholipid; and a PEGylated lipid. Such lipid nanoparticles have been found to be particularly effective at encapsulating and delivering nucleic acids, such as mRNA. The phospholipid which may be comprised in the lipid nanoparticle analyte may be a glycerophospholipid or a phosphosphingolipid. For instance, the phospholipid may be a glycerophospholipid selected from: « phosphatidylcholines, such as distearoylphosphatidylcholine (DSPC), l-palmitoyl-2-oleoylphosphatidylcholine (POPC), dipalmitoylphosphatidylcholine (DPPC), 1,2-dipentadecanoyl-sn-g1yc ero-3-pho s pho cho1i ne a n d d i my ri s to y1phos ph a t i d y1 cho1i ne (DMPC); ® phosphatidylethanolamines, such as distearoylphosphatidylethanolamine (DS PE), dimy ris toy1p ho s ph atidy1ethanolami ne (DMPE), and dipalmitoylphosphatidylethanolamine (DPPE); * phosphatidylserines such as 1,2-dimyristoyl-sn-glycero-3-phospho-L-serine (DMPS); * phosphatidylinositols; * phosphatidylglycerols such as 1,2-Dipalmitoyl-sn-glycero-3-phosphoglycerol (DPPG); and * phosphatidic acids such as 1,2-dimyristoyl-sn-glycero-3-phosphate (DMPAj , The phosphospingolipid may be selected from ceramide p ho s ph o ry1cho1ines, ceramide pho s ph o r y1etha no1ami nes, and cerami de phosphory11ipids . A PEGylated lipid refers to a conjugate molecule that comprise a PEG (polyethylene glycol) moiety attached to a lipid moiety. Thus, the PEGylated lipid which may be comprised in the lipid nanoparticle analyte may be selected from: PEG-modified phosphatidylethanol amines such as DSPE-PEG; PEG-modified phosphatidic acids; PEG-modified ceramides; PEG-modified dialkylamines; PEG-modified diacylglycerols; and PEG-modified dialkylglycerols. Examples of suitable PEGylated lipids 'which may be of use in the methods described herein include PEG-c-DOMG, PEG-DMG, PEG-DLPE, PEG-DMPE, PEG-DPPC, PEG-DSPE, PEG-DSG, PEG-DAG, PEG-DPPE, PEG-c-DMA, PEG-dipaImetoleyl, PEG-dioleyl, PEG-distearyl, and ADC-0159. The structural lipid that may be present in the lipid nanoparticle formulations or precursor formulations may be selected from sterols such as cholesterol and derivatives thereof, fecosterol, sisosterol, campesterol, stigmasterol, brassicasterol, ergosterol, desmosterol, zymosterol, and lanosterol. A cationic and / or ionisable lipid refers to a lipid which either permanently holds a net positive charge (a cationic lipid) or which remains neutral at physiological pH but can become protonated at lower pH to render the lipid charged (an ionisable lipid). A cationic lipid comprises a cationic group. For instance, the cationic lipid may comprise one or more quaternary amines. Examples of suitable cationic lipids that may be present in the lipid nanoparticle analytes include: monovalent aliphatic lipids such as N-(1-(2,3-dioleyloxy)propyl)~N,N,N-trimethyl ammonium chloride (DOTMA), 1,2-Dioleoyl-3-trimethylammonium propane (DOTAP), (3- dimyristyloxypropyl)(dimethyl)(hydroxyethyl)ammonium (DMRIE) and 1[2- (9 (Z) -octadecenoyloxy) ethyl] -2-- (8 (Z) -heptadecenyl) -3- (2-hydroxyethyl)imidazolinium chloride (DOTIM); multivalent aliphatic lipids such as dioctadecylamidoglycylspermine (DOGS); and cationic lipid derivatives such as 3^-(N-(N',N’-dimethylaminoethane)-carbamoyl) cholesterol (DC-Chol). Preferably, the cationic and / or ionisable lipid is an ionisable lipid. Typically, an ionisable lipid comprises an ionisable headgroup which is able to be deprotonated at physiological pH (around pH 7.4), but which becomes protonated at lower pH values (e.g. pH <6). Suitable ionisable headgroups include tertiary amines, guanidines, imidazoles, and piperazines, in particular a tertiary amine. The ionisable headgroup is connected to one or more alkyl tails, typically via a linker group. These alkyl tails may be branched, saturated or unsaturated, and are typically in the range of 8 to 18 carbons in length. The linker group may be a Ci-Cio alkyl terminating in an ester, an ether, an amide, a thioether, a carbonate, a disulfane, or a urethane. The structures of ionisable lipids, which may be present in the lipid nanoparticle analytes tested in the methods described herein, are described in WO 2023 / 009421 Al. Exemplary ionisable lipids include ALC-0315, SM-102, DLin-MC3-DMA, C12-200, 5A2SC8, 9A1P9, 306-O12B-3, YSK05, 5A2-SCB, TCL053, CKK-E12. A non-phospholipid surfactant refers to a surfactant which is not a phospholipid. Examples of suitable non-phospholipid surfactants that may be present in the lipid nanoparticle analytes include: • non-ionic surfactants such as fatty alcohol ethoxylates, alkylphenol ethoxylates, alkyl polyglucosides, fatty acid esters of glycerol, a cocamide MEA, a cocamide DEA, and fatty acid esters of sorbitol such as polysorbates; • anionic surfactants, such as alkyl sulfates or alkyl ether sulfates, alkyl sulfonates or alkyl ether sulfonates, and alkyl carboxylates or alkyl ether carboxylates; • amphoteric: surfactants such as alkylbetaines, alkylami ne oxides, alkylamidobetaines, alkylamidoamine oxides, and sulfobetaines. Precursor lipid nanoparticle formulations Advantageously, the methods of the present disclosure may also be used to assess the biophysical characteristics of precursor lipid nanoparticle formulations. Precursor lipid nanoparticle formulations do not comprise an assembled lipid nanoparticle, but do comprise a mixture of constituent biomolecules of a lipid nanoparticle, suitable for forming at least part of an assembled lipid nanoparticle. For instance, it can be useful to assess the potential developability of a precursor formulation while its constituent biomolecules are still in a solubilised state, without having to perform the synthetic steps to assemble the precursor formulation into assembled lipid nanoparticles. A precursor lipid formulation is able to form a stable lipid nanoparticle formulation (for instance, upon introduction to an aqueous solvent), the nanoparticles of which do not readily coalesce. As such, it is preferred that the precursor lipid nanoparticle formulation comprises an amphiphilic compound, such as a phospholipid or a surfactant. In particular, the precursor lipid nanoparticle formulation may comprise a mixture of two or more of: a phospholipid; a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid; a nonphospholipid surfactant; a tri--, di-, or mono-glyceride; a fatty acid; and a wax. Preferably, the precursor lipid nanoparticle formulation comprises one or more of a PEGylated lipid and a cationic and / or ionisable lipid. For instance, where the precursor lipid nanoparticle formulation comprises a tri-, di-, or mono-glyceride, a fatty acid, or a wax, it is preferred that the precursor lipid nanoparticle formulation also comprises a PEGylated lipid or a cationic and / or ionisable linid. For instance, the precursor lipid nanoparticle formulation may comprise a mixture of two or more, preferably three or more, and more preferably four or more of: a phospholipid; a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid. In particular, it is preferred that the precursor lipid nanoparticle formulation comprises a phospholipid, a structural lipid, a PEGylated lipid and a cationic and / or ionisable lipid. Typically, the sample comprising a precursor lipid nanoparticle formulation will also comprise; a solvent suitable for dissolving the constituent biomolecules of the precursor formulation. In some embodiments, the sample consists of the precursor lipid nanoparticle formulation and a solvent system, preferably ’wherein the precursor lipid nanoparticle formulation consists of a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid; and a phospho1ipid. In particular, it is preferred that the precursor lipid nanoparticle formulation is dissolved in an organic solvent or an organic solvent system. Any solvent suitable for dissolving a lipid nanoparticle formulation may be used. Examples of suitable solvents and solvent systems include chloroform, dichloromethane, methanol, ethanol, acetone, acetonitrile, hexane, methanol-chloroform mixtures, and mixtures thereof. Where the sample comprises a lipid nanoparticle analyte: which is a lipid nanoparticle formulation or a precursor formulation, the lipid nanoparticle analyte may comprise a cationic and / or ionisable lipid in an amount of at least 30 moll, preferably at least 40 moll, and more; preferably at least 5 0 moll of the; lipid nanoparticle analyte. The lipid nanoparticle analyte may comprise; a cationic and / or ionisable lipid in an amount of up to 7 0 moll, preferably up to 60 moll, and more preferably up to 55 moll of the candidate composition. Thus, the lipid nanoparticle analyte may comprise a cationic and / or ionisable lipid in an amount in the range of from 30 to 70 moll, preferably from 40 to 60 moll, and more preferably from 45 to.' 55 moll of the lipid nanoparticle analyte. Likewise, the lipid nanoparticle analyte may comprise a phospholipid in an amount of at least 0.1 moll, preferably at least 0.5 moll, and more; preferably at least 1 mol% of the; lipid nanoparticle analyte. The lipid nanoparticle analyte may comprise a phospholipid in an amount of up to 20 moll, preferably up to 15 moll, and more preferably up to 12.5 moll of the lipid nanoparticle analyte. Thus, the lipid nanoparticle analyte may comprise a phospholipid in an amount in the range of from. 0.1 to 20 moll, preferably from 0.5 to 15 moll, and more; preferably from 1 to 12.5 moll of the lipid nanoparticle analyte. The lipid nanoparticle analyte may comprise a structural lipid in an amount of at least 20 mo 1%, preferably at least 30 moll, and more preferably at least 35 mol% of the lipid, nanoparticle analyte. The lipid nanoparticle analyte may comprise a structural lipid in an amount of up to 60 mol%, preferably up to 55 mol%, and more preferably up to 50 mol% of the lipid nanoparticle analyte. Thus, the lipid nanoparticle analyte may comprise a structural lipid in an amount in the range of from 20 to 60 mol%, preferably from. 30 to 55 mol%, and more preferably from 35 to 50 mol% of the lipid nanoparticle analyte. The lipid nanoparticle analyte may comprise a PEGylated lipid in an amount of at least. 0.1 mol%, preferably at. least. 0.5 mol%, and more preferably at. least 1 mol% of the lipid nanoparticle analyte. The lipid nanoparti.cle analyte may comprise a PEGylated lipid in an amount of up to 20 mol%, preferably up to 15 mol%, and more preferably up to 12.5 mol% of the lipid nanoparticle analyte. Thus, the lipid nanoparticle analyte may comprise a PEGylated lipid in an amount in the range of from 0.1 to 20 mol%, preferably from 0.5 to 15 mol%, and more preferably from 1 to 12.5 mol% of the lipid nanoparticle analyte. Where the lipid nanoparticle formulation or precursor lipid nanoparticle formulation comprises a structural lipid (A), a PEGylated lipid (B), a cationic and / or ionisable lipid (C>, and a phospholipid (D) , the molar ratio of constituents A:B:C:D in the precursor lipid nanoparticle formulation may be in the range of from 20-60:0.1-20:3070:0.1--20, preferably from 30--55:0.5--15:4 0--60:0.5--15, and more preferably from 35--50:1--12.5:45-55:1--12.5. The lipid nanoparticle formulation or precursor formulation may further comprise a therapeutic compound, for instance a nucleic acid such as mRNA., or a small molecule drug which may be a lipophilic, small molecule drug. For instance, where the sample comprises a lipid nanoparticle formulation, the therapeutic compound may be encapsulated within the assembled nanoparticles. The lipid nanoparticle formulation or precursor formulation may also comprise a protein, such as a targeting protein that facilitates targeted delivery of the lipid nanoparticle to a particular site or target when administered. For instance, where the sample comprises a lipid nanoparticle formulation, the protein nay be located on an external surface of the nanoparticle, such as in the form of a protein corona . Constituent biomo1ecu1es The methods of the present disclosure nay be used to assess the biophysical characteristics of individual lipid compounds that can be used in lipid nanoparticle formulations. Thus, the sample may comprise a lipid nanoparticle analyte which is a constituent biomolecule of a lipid nanoparticle. The constituent biomolecule may be selected from: cholesterol or a derivative thereof; a cationic and / or ionisable lipid; a phospholipid; a PEGylated lipid; or a surfactant. It will be appreciated that the definitions of these species may be the same as defined above. It will be appreciated that there is no particular limit on the nature of the lipid nanoparticle formulations, precursor lipid nanoparticle formulations or constituent biomolecules that can be used in the methods described herein, or compositions comprising them. The methods of the present disclosure may be used to probe the biophysical characteristics of known lipid nanoparticle analytes, or those under clinical development. This can in turn represent a means for determining the viability of a candidate lipid nanoparticle formulation, precursor formulation or constituent biomolecule under development from an earlier stage, based on a comparison of biophysical characteristics of the candidate versus a target lipid nanoparticle analyte (e.g. a commercially available lipid nanoparticle formulation with a proven therapeutic efficacy). In some embodiments, the biophysical characterisation provided by the methods described herein comprises one or more properties selected from: compressibility, saturation pressure, surface activity, viscoelasticity, viscosity, stability in a particular environment, hydrophobicity, binding affinity to a particular target, solubility in a particular solvent, surface charge, specific gravity, phase transition, and polydispersity index. Preferably, the biophysical characterisation comprises a plurality of data values suitable for identifying parameters indicative of at least one phase transition of the analyte for example the data may be suitable for defining the parameters at which such a phase transition occurs and may be suitable for providing at least part of a phase diagram or for providing a polydispersity index for the lipid nanoparticle analyte. When the lipid nanoparticle analyte undergoes a phase transition, it may result in a deviation from linearity in an aspect of the wave data generated. Identifying phase transitions in the lipid nanoparticle analyte can be used to construct a phase diagram for the lipid nanoparticle analyte under the conditions of the method. Thus, the methods described herein may comprise: identifying non-linearity in the wave data response, preferably wherein the non-linearity is not associated with the critical micelle concentration of the lipid nanoparticle analyte. For instance, the methods of the present disclosure may be used to analyse the physicochemical tropism of a lipid nanoparticle analyte, based on certain biophysical characteristics of the lipid nanoparticle analyte. Of particular relevance to this analysis may be the interactions and binding affinity of the lipid nanoparticle analyte with certain biological species, such as blood cells, platelets and plasma proteins. For instance, serum proteins such as albumins, globulins, thrombin, fibrinogens, and fibrin may be of particular interest. As such, to generate wave data describing the interaction betvjeen the lipid nanoparticle analyte and the biological species of interest, the biological species may be comprised in the reservoir medium, comprised in a thin film upon the surface of the reservoir medium, or provided as a constituent of a separate sample which is deposited onto a thin film formed by deposition of a lipid nanoparticle analyte . The possibility of using biophysical characterisation of a biomolecule to predict clinical outcomes has been established. For instance, in the field of small molecule drug discovery, Lipinski’s "rule of five" (whereby factors such as total number of hydrogen bond donors and acceptors, molecular weight and octanol-water partition coefficient are used to predict a drug's pharmacokinetics’ has seen widespread applicability. It has also been reported that developability of antibodies (i.e. their suitability for becoming a safe and efficacious drug) can be correlated to certain biophysical characteristics such as polyspecificity, melting temperature, aggregation tendency and hydrophobicity (see e.g. Jain et al., "Biophysical properties of the clinical-stage antibody landscape", Proc. Natl. Acad. Sci . USA, 2017;114(5) : pages 944-9, and Bailly et al., "Predicting Antibody Developability Profiles Through Early Stage Discovery Screening", Mabs, 2020, VOL. 12, No.l, el743053). Physicochemical tropism in lipid nanoparticles, whereby biophysical characteristics are tailored to target lipid nanoparticles to specific cell or organ types, has also been described (see e.g. Omo-Lamai et al., "Physicochemical Targeting of Lipid Nanoparticles to the Lungs induces Clotting: Mechanisms and Solutions", bioRxiv, July 25 2023.) Thus, the method may further comprise a step iii) of providing a metric of developability for the lipid nanoparticle analyte, such as manufacturability or safety profile. For example, the metric of developability may be based on a plurality of biophysical characteristics, said plurality of biophysical characteristics each being derived from one or more separable contributions of the wave data. The metric of developability may be selected from any one or more: in vivo performance such as biodistribution, reduction in mortality or morbidity, and protein expression level; in vitro performance such as transfection efficacy; encapsulation efficiency; polydispersity index; lipid nanoparticle size; and stability. As will be appreciated, the methods described herein may also be used to assess any potential level of contamination across a series of lipid nanoparticle analyte samples, and the extent of the contamination may be assessed based on the effect of the contamination on the prevailing biophysical characteristics. Sample comprising the lipid nanoparticle analyte The methods of the present invention involve the generation of wave data in a liquid system. - employing a liquid sample in which a lipid nanoparticle analyte is incorporated into a liquid carrier. As will be appreciated, the particular nature and composition of the lipid nanoparticle analyte will impact the most suitable form the sample may take - i.e. the particular way in which the lipid nanoparticle analyte is disposed within a liquid carrier. For example, the lipid nanoparticle analyte may be incorporated in solution, as a suspension, or as part of an emulsion, depending on the miscibility of the lipid nanoparticle analyte in a given liquid carrier, or the melting point of the lipid nanoparticle analyte. The skilled person is readily able to select a liquid carrier based on the particular class of lipid nanoparticle analyte being investigated. For instance, where the lipid nanoparticle analyte comprises a lipid nanoparticle formulation, an aqueous liquid carrier is generally most appropriate in order for the lipid nanoparticles therein to retain their nanoparticulate form. Suitable aqueous liquid carriers include water (e.g. deionized water), a glycerol-water mixture, saline {e.g. phosphate buffered saline (PBS)) or medical buffer solutions {e.g. tris(hydroxymethyl) aminomethane buffer (THAM)). As already mentioned, the lipid nanoparticle formulation may be in the form of an emulsion, specifically an oil-in-water microemulsion or a nanoemulsion, which may be generated by known methods such as high pressure homogenisation. For instance, such an emulsion may be readily prepared by mixing the lipid nanoparticle formulation with an organic solvent, before injecting it into an aqueous solvent {e.g. deionized water) , agitating the mixture {e.g. using a high speed, stirrer) and applying heat {e.g. from 30 to 60 °C) under pressure {e.g. in the range of 500 to 1500 bar), and optionally repeating the agitation, until an emulsion is formed. Typically, tne sample comprising a precursor lipid nanoparticle formulation will also comprise a solvent suitable for dissolving the constituent biomolecules thereof. In particular, it is preferred that the precursor lipid nanoparticle formulation is dissolved in an organic solvent or solvent ; dissolving a lipid nanoparticle suitable organic solvents and dichloromethane, alcohols (such acetone, acetonitrile, hexane, methanol-chloroform mixtures . ystem. Any solvent suitable for formulation may be used. Examples of solvent systems include chloroform, as methanol, ethanol and 2--propanol) , and combinations thereof, such as The lipid nanoparticle analyte may be at substantially any concentration within the sample. However, the concentration is typically tailored to the surface area of the reservoir medium, such that deposition of a single sample droplet does not fully saturate 1:.ne surface with the 1 ipid nanoparticle analyte, allowing a series of' droplets (and analysis of the resultant waves) to be performed before saturation of the surface. For instance, lipid nanoparticle analytes may be provided at a concentration in the range of from 0.01 mg / ml to 10 mg / ml, preferably in the range of from 0.1 to 1 mg / ml. Reservoir Medium As described briefly above, the reservoir medium employed in the methods described herein is a stable liquid (over the timescale of the experiment), typically a solution or suspension, through which surface waves can be propagated upon deposition of a sample therein. For ease of operation and reproducibility, simple aqueous solutions are typically employed in the method described herein. For example, deionized water or a brine / alkali metal halide salt solution (e.g. NaCl) may be used, the latter being of potential benefit in encouraging thin film retention / formation at the surface of the liquid reservoir, where present or desired. Surfactants may also be included in the liquid medium for the purpose of forming a thin film at the liquid reservoir surface as a means to modify or optimise the waveform response following sample deposition, if desired. Examples of suitable surfactants i n c1ude anionic surfactants (e-g. those containing a sulfate or sulfonate groups), In some embodiments, the reservoir medium may comprise an excipient, such as would be suitable for use in a pharmaceutical composition comprising a lipid nanoparticle formulation. Excipients are often present in lipid nanoparticle-based therapeutics in order to provide improved properties such as stability, solubility of active ingredients, cryoprotection, and viscosity modification. By providing an excipient in the reservoir medium, the interaction between the lipid nanoparticle analyte and the excipient, can be examined using the wave data generated, which can be used to e.g. assess the compatibility of the excipient and the lipid nanoparticle analyte. Suitable excipients include: emulsifiers such as polysorbates and poloxamers; salts such as sodium chloride and sodium phosphate; saccharides such as sucrose, lactose, trehalose and glucose; sugar alcohols such as mannitol, and sorbitol; amino acids such as histidine, methionine, arginine, proline, glutamate and glycine; acids such as hydrochloric acid and citric acid; buffering agents such as sodium citrate and sodium succinate; and preservatives such as benzalkonium ch1oride. The reservoir medium may also comprise a therapeutic compound, such as a lipophilic small molecule drug or an mRNA, or a protein. As these may also be components of eventual clinical lipid nanoparticle formulations, their presence in the reservoir medium can enable the interactions between these components and the lipid nanoparticle analyte to be studied. Biophysical characterisation The methods described herein are capable of providing biophysical characterisation of a lipid nanoparticle analyte based on wave data generated in connection therewith. These characteristics include several different categories of characteristics, including those that directly affect the droplet formation and spread in the liquid system (i.e. relating to fluid dynamic parameters’, as well as properties of compressible fluids, such as compressibility, bulk (second) viscosity, thermal conductivity, and heat capacity that play a role in wave propagation. In addition, there are enthalpies of interaction between the droplet and the liquid in the reservoir which directly contribute energy into the wave, and which are therefore derivable from the wave data, particularly enthalpies related to interfaces such as surface charge of the droplet, partition coefficient (hydrophobicity), and surface pKa. Kinetics and timescales of these interactions are also accessible (and are some of the above properties are time timescale dependent properties). Timescales may, for instance, carry information about the conformational changes in a system. Short timescales (i.e. <microsecond) may correspond to enthalpy of intramolecular conformation changes; medium timescales (i.e. microsecond to millisecond) may correspond to collective intramolecular changes in large molecules (e.g. allosteric changes); and long timescales (i.e. >millisecond) may be intramolecular as well as intermolecular collective changes (e.g. in a fluid to gel phase change). The wave data is also capable of assessing indirect properties of the lipid nanoparticle analyte, based on comparison to the wave data of a notional standard lipid nanoparticle analyte, where incremental compositional changes away from the composition of the notational standard can give indirect insights into biophysical properties, such as freezing point depression, vapor pressure (volatility), osmotic pressure, boiling point elevation, critical micel1ization concentration, and surface pressure. Thus, in another aspect, there is provided a method for determining enthalpies of interaction for a sample of a lipid nanoparticle analyte with a liquid system, said method comprising the steps of: i) obtaining wave data for a sample based on physical measurements of a wave generated, in a liquid system, by providing contact, between a droplet of the sample and the liquid system, wherein the wave comprises modes encoding characteristics of the interaction between the sample and the liquid system, wherein the sample comprises the lipid nanoparticle analyte, and wherein the lipid nanoparticle analyte comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation; i i) det ermi ni n g entha1p i es, k i n etic s an d / o r timescales of interaction of the sample and the liquid system based on the wave data . In yet another aspect, there is provided a method of identifying a lipid nanoparticle analyte with comparable enthalpies, kinetics and / or timescales of interaction to a target lipid nanoparticle analyte, said method comprising the steps of: (i) obtaining wave data for a sample based on physical measurements of a wave generated, in a liquid system, by providing contact between a droplet of the sample and the liquid system, wherein the wave comprises modes encoding characteristics of the interaction between the sample and the liquid system, wherein the sample comprises the lipid nanoparticle analyte, and wherein the lipid nanoparticle analyte comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation; (ii) providing a comparison of the wave data obtained from interaction of the sample with the surface of the liquid system with wave: data associated with the target lipid nanoparticle analyte; optionally repeating steps i) to ii) for a different lipid nanoparticle analyte until a lipid nanoparticle analyte is identified having comparable values of at least one form of enthalpy of interaction, indicative of comparable biophysical properties of the lipid nanoparticle analyte and the target lipid nanoparticle analyte. The present invention will now be illustrated by way of the following examples and with reference to the figures. General method for sample preparation, deposition and wave data gathering 1. Obtain a lipid nanoparticle analyte for testing and prepare a solution, suspension, or emulsion of the lipid nanoparticle analyte in a liquid carrier at a w / v concentration of 0.1 mg / ml to 0.5 mg / ml in a sample tube (e.g. a LowBind Eppendorf tube); 2. Prepare a reservoir liquid medium and optionally degas under va cuum; 3. Incorporate the reservoir medium into a reservoir (e.g. a Langmuir trough) into which the sample; may be deposited; 4. Deposit a set volume (e.g. 1 to 100 pl) in the form of one or more droplets of the sample into the reservoir using a droplet provider (e.g. a single-channel pipette) from a fixed location above the air-water interface of the reservoir (at a height of 2 to 5 mm above the air-water interface) and at a fixed volumetric flow rate (e.g. 10 to 200 pl / minute) affording a fixed droplet frequency - i.e. the number of droplets produced per second -(e.g. 0.1 to 1000 droplets / s); 5. Measure the response of the reservoir air-water interface to the first droplet, and any successive droplets, dispensed into the reservoir from the fixed location, using an apparatus incorporating: i) a light beam optics comprising a source of polarised light arranged to illuminate an area of liquid at the surface of the reservoir with a beam having a selected angle of incidence a; ii) a light collector arranged to receive a beam of light after reflection by the area of the thin film, and iii) a detector configured to sense parameters of the light received from the light collector to provide signals to a wave measurement module 17 (for example, an apparatus as shown in Figure 2 and described in further detail herein). As will be appreciated, deposition of the sample may be automated and a continuous flow rate pump system with an autosampler and droplet dispenser may suitably be used, for such purposes, for instance an HPLC (without chromatography columns installed) with an autosampler and a dispenser. An Opentrons (RTM) system, as described e.g. in PCT application PCT / GB2024 / 050857, may also be used according to the ways described therein to provide the sample droplets, provided it is set up to.' tolerate the typical solvents used for lipid nanoparticle analytes . It will be appreciated in the context of the present disclosure that appropriate experimental controls should be applied, for example to ensure that the waves caused by the different droplets in a series are comparable with each another. Control of droplet size and flow rate is one such factor, but others will be apparent to the skilled person in the context of the present disclosure. A dispenser and / or autosampler for an HPLC column may offer one way to usefully control droplet size and delivery rate but any appropriate dispenser may be In addition, the wave data may be conditioned and cleaned prior to analysis by routine methods. Such methods may include removing anomalies from the wave data, such as noisy measurements or measurements affected by factors external to the instrument used to perform the measurement -- which may arise from the laboratory or other environment in which it is deployed. Example 1: Constituent Biomolecule Evaluation The properties of a number of constituent biomolecules were investigated following the general method described above. Samples of i) 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), ii) di st earoylphosphatidyl cho line (DS PC) , iii) dipalmitoyl--phosphatidylcholine (DPPC), iv) 1,2-dipentadecanoyl-sn-glycero-3-phosphocholine ¢15:0 PC), and v) 1,2-dioleoyl-sn--glycero--3- phosphoethanolamine; (DOPE) were prepared and tested. Each of these constituent biomolecules were purchased from Avanti Polar Lipids. Samples containing the different constituent biomolecules were prepared in the same manner and at the same concentration - 0.2 mg / mL. For each constituent biomolecule investigated, a sample was formed by adding an appropriate amount of constituent biomolecule to the chloroform solvent, to produce the 0.2 mg / ml solutions. The chloroform contained a small amount of ethanol (around 0.5---1%) as stabiliser. 1.5 ml of each solution 'was then aliquotecl into separate Eppendorf tubes, allowed to equilibrate to room temperature over a 10-minute time period and placed in an HPLC system with an autosampler and dispenser for dispensing the sample droplets. The reservoir medium for use in testing in these experiments was deionised water for each sample. The reservoirs for use in these experiments were each of dimensions 12 0 mm x 8 0 mm and filled with the deionised water with a liquid height alignment of the reservoir being performed optically (using trigger laser intensity, monitored as reservoir medium is extracted from the reservoir until a fixed trigger laser intensity is reached) before the start of each experiment to ensure consistency. The HPLC system with a dispenser was used for sample deposition and an apparatus substantially as shown in Figure 2 and described herein was used for obtaining surface wave dan a . The HPLC system and dispenser was configured to give a sample deposition volume of 80-100 uL (as a series of droplets), from a height of less than 1mm, and to include an isopropanol and water wash between each constituent biomolecule sample testing. During the experiment, the sample deposited by the series of droplets was allowed to accumulate on the surface of the reservoir, eventually forming a thin film of the constituent biomolecule in the sample. At least one of the droplets of the series was deposited on the formed thin film. The surface pressure of the film at the air-water interface was measured with a Whilhelmy plate, to independently verify the increase in surface pressure as sample deposition occurs, before eventual saturation of the surface with the sample forming a thin film. Surface wave data generated from these experiments was subsequently used to generate a biophysical characterisation of each of the c onstituent biomolecules, i n c o r p o r a ting p r i n c i p a 1 c o mp o n e n t a n al. y sis (PCA) . The results, which are represented visually in Figure 8, demonstrate that the present invention allows a unique signature for different constituent biomolecules, or compositions thereof, to be acquired, representing a combination of biophysical properties of the constituent biomolecule sample. This can be seen in the plot in Figure 8A, which shows distinct clustering for individual constituent biomolecule samples. For instance, the saturated phospholipids DMPC, 15:0 PC, DSPC, DPPC and DSPC form clusters more closely to each other-compared to the unsaturated lipid, DOPE. Separation between particular constituent biomolecules was also able to be tuned by varying the principal components being compared. These data may be used, to populate a library of such signatures for comparison purposes, and / or used as means to compare against corresponding data of a target biomolecule to determine whether the biomolecule investigated has comparable biophysical properties. PCA was found, to reveal a direct correlation between the measured, signatures for the constituent biomolecule samples of these experiments and the measured surface pressure. Using PCA, it was also possible to determine the expected phase transitions for 15:0 PC and DPPC, with these: phase transitions occurring at the expected surface pressures - the phase transitions are shown in Figure 8B, as the nonlinear deviations from the linear portion of the sigmoidal curves. The other constituent biomolecules do not have any phase transitions under the experimental conditions. Phase transition data derived from the generated wave data can be used to provide phase diagrams for a particular lipid nanoparticle analyte. Phase states and phase transitions of lipid nanoparticle formulations is believed to be influential in providing effective clinical outcomes. Example 2: Precursor Lipid Nanoparticle Formulation Evaluation The properties of a number of different precursor lipid nanoparticle formulations were investigated following the general method described 5 above. Samples of different precursor lipid nanoparticle formulations were prepared and tested. The constituent biomolecules of each sample are set out in Tables la and lb, below: Formulation No. Pho spholipid (DSPC); mol% PEGylated lipid (DSPE-PEG2k); mol% Structural lipid (Cholesterol); mol% lonisable lipid (ALC-0315) ; iaol% 1 10.0 1.5 38.5 5 0 2 8.0 1.5 Jb . 5 52 8 5 . 0 1.5 3 8,5 55 4 1.0 1.5 3 8,5 5 9 0 8.0 3.5 3 8,5 o U 6 5.0 6.5 38.5 50 / 1.0 10.5 38.5 5 0 8 8.0 1.5 4 0.5 5 0 9 5.0 1.5 4 9 . 5 5 0 10 1.0 1.5 4 7.5 u. 'J Formulation § Phospholipid No. | (DPPC); mol% PEGylated lipid (DSPE-PEG2k); mol% Structural lipid (Cholesterol) ; iaol% lonisable lipid (DODMA) ; 37101 11 | 10. 0 1.5 3 8.5 5 0 12 | 8.0 1.5 38.5 c2 ; i-.1. o 1.5 38.5 5 5 14 |1.0 1.5 38.5 5 9 ! 5 8.0 3.5 38.5 50 16 5.0 6.5 38.5 5 0 1 / 1.0 10.5 38.5 5 0 18 8.0 1.5 4 0.5 5 0 1 5.0 1.5 4 3.5 u. 'J 2 0 1.0 1.5 47.5 5 0 Formulations 1--20 were formed by providing each of the constituent biomolecules (sourced from Avanti Polar Lipids) at the same desired concentration 0.2 mg / m1, by di1ut i on with ethanol-stabilised chloroform. The final precursor lipid nanoparticle formulations were then formed by mixing together each of the constituent biomolecule samples in the appropriate ratios, as described above. The reservoir medium for use in testing formulations 1-20 was a: stock solution of deionised water. The reservoirs for use in these experiments were of dimensions 40 mm x 80mm and filled with the stock aqueous solution, a liquid height alignment of the reservoir was performed optically (using trigger laser intensity, monitored as reservoir medium is extracted from the reservoir until a fixed trigger laser intensity is reached) before the start of each experiment to ensure consistency. The HPLC system with a dispenser ’was used for sample deposition and an apparatus substantially as shown in Figure 2 and described herein was used for obtaining surface wave data. The HPLC system with dispenser was set to give a sample deposition volume of 80-100 pL (as a series of droplets), from a height of less than 1 mm, and to include an isopropanol and water wash between each precursor lipid nanoparticle sample testing. During the experiment, the surface pressure of the film at the air-water interface was measured with a Whilhelmy plate, to independently verify the increase in surface pressure as sample deposition occurs, before eventual saturation of the surface with the sample. Surface wave data generated from these experiments was subsequently used to generate a biophysical characterisation for each of the precursor lipid nanoparticle formulations. Figure 9A is a hierarchical clustering heat map using the surface pressure response of the sample onto the reservoir surface. In this view of the selected data, there are 3 distinct clusters that map to high ionisable lipid, nigh PEGylated lipid concentration and varying structural lipid concentration. For example, the upper left cluster shows that the most dissimilar formulations are formulation 14 and formulation 7. These two formulations are quite different as they belong to the two different classes, but vary most significantly in their PEGylated lipid concentration. Formulation 14 has a PEGylated lipid concentration of 1.5% wliile formulation 7 has a PEGylated lipid concentration of 10.5%. A comparison between e.g. formulation 9 and formulation 19 also reveals that when the relative ratios of the components are the same but the phospholipid and ionisable lipid differ, a different response is still observable. Figures 9B-9D show the results of principle component analysis (PGA) from wave data obtained for the precursor lipid nanoparticle formulations. Figure 9B contains the data for two sets of formulations: formulations 1---4 and formulations 11--14, these two sets differing only in the structure of the constituent phospholipid and ionisable lipid components. As can be seen from Figure 9B, formulations 1-4 and 11-14 form distinct, separable clusters according to the constituent biomolecules. Like’wise, Figure 9C shows that formulations 5-7 and 15-17, which again differ only in the constituent phospholipid and ionisable lipid structures, can be separated according to their constituent biomolecules. Within each of these two sets, a trend is also observable based on the amount of PEGylated lipid present in the c omp o s i t i o n s. Figure 9D depicts formulations 8-10 and 18-20, with the two sets again being separable according to their constituent biomolecules. A trend may also be observed within the set of formulations 8-10, stratifying the formulations according to the amount of cholesterol present therein. This further demonstrates the ability of the methods of the present disclosure to separate lipid nanoparticle formulations, both according to their constituent biomolecule structure and amount. The predictive ability of the present methods have been further demonstrated in e.g. antibodies and glyceride oils. For instance, for glyceride oil samples the correlations in the PCA were validated by correlations between independently determined biophysical and sensory properties of the oils, including correlations between density, viscosity. Iodine value, molecular weight, mono-unsaturated fatty acid content (MUFA), poly-unsaturated fatty acid content (PUFA) and surface tension of the oils. An aspect of the disclosure provides an apparatus for biophysical characterisation of a lipid nanoparticle analyte, said apparatus comprising : an analytical instrument configured and arranged to perform physical measurements of a wave in a liquid system, and to obtain, from said measurement, wave data for the sample based on physical measurements of the wave generated, in the liquid system, by providing contact between the liquid system and a droplet of a sample comprising the lipid nanoparticle analyte, wherein the wave comprises modes encoding characteristics of the interaction between the sample and the liquid system, and the apparatus comprises a data processor configured to provide a biophysical characterisation of the lipid nanoparticle analyte based on the wave data. The analytical instrument may comprise a droplet provider configured to contact the liquid with a droplet of the sample. The analytical instrument may comprise a wave measurement module configured to provide the wave data based on measurements of the wave. The measurements of the wave may be based on interaction of a light beam, such as a laser with the liquid system. The analytical instrument nay comprise a trough holding the liquid system and the liquid system may comprise a thin film. The data processor may be connected to receive the wave data from the analytical instrument. The data processor may also be connected to a data store, such as a digital memory, storing association data. The analytical instrument may be configured to generate wave data comprising a representation of a plurality of said waves generated in the liquid system, each of those waves corresponding to contact of a different one of a plurality of droplets with the liquid system. The data processor may be configured to provide the biophysical characterisation by identifying a plurality of separable contributions to the variance of the wave data. For example, the data processor may be coupled to communicate with data storage storing association data defining a relation between at least one of the separable contributions and a biophysical characteristic. The data processor may be configured to determine data indicating the biophysical characteristic based on the separable contributions and the association data. It is contemplated that the apparatus described in this paragraph is to be configured to perform any one of the methods described or claimed herein. The present disclosure provides a process for manufacturing a product comprising or made from, a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, the process comprising: obtaining wave data: for a sample based on physical measurements of a wave generated, in a liquid system, by providing contact between a droplet of the sample and the liquid system, wherein the sample comprises a lipid nanoparticle analyte which comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation, and the wave comprises modes encoding characteristics of the interaction between the sample and the liquid system, providing a biophysical characterisation of the lipid nanoparticle analyte based on the wave data; and ma n u f a c t u r i n g t h e p r o du c t. The biophysical characterisation nay be provided according to any one of the methods described or claimed herein. The manufacturing may be done according to the biophysical characterisation. The lipid nanoparticle analyte may comprise; a part or all of the product, or an ingredient or other constituent part of the product. For instance, where the lipid nanoparticle analyte comprises a lipid nanoparticle formulation, the product may be a vaccine formulation comprising the lipid nanoparticle formulation and suitable excipients. Likewise, where the lipid nanoparticle analyte comprises a precursor lipid nanoparticle formulation, the product may be an assembled lipid nanoparticle formulation (e.g. a vaccine formulation) that comprises the constituent compounds of the precursor formulation. The biophysical characterisation may provide information for use in making the product or for use with the product. For example, the information may be used to verify quality or purity or to adjust a step in the manufacturing process. For example, the product can then be made by making use of said information, such as to control or to verify the quality of said manufacturing process. As another example, the lipid nanoparticle formulation or precursor formulation may be adjusted based on the information. This may comprise varying a ratio of excipients or constituents of the formulation or precursor formulation . Where the lipid nanoparticle analyte is a constituent biomolecule or a precursor lipid nanoparticle formulation, manufacturing the product may comprise performing steps to provide an assembled lipid nanoparticle formulation. Methods of forming lipid nanoparticles are known to the skilled person, and include e.g. high-pressure homogenisation. Optionally, manufacturing the product may comprise: combining the lipid nanoparticle analyte with one or more further lipids, before forming said lipid nanoparticles. The addition of excipients, as described herein, may also form part of the manufacture of the product. Any feature of any one of the examples disclosed, herein may be combined with any selected features of any of the other examples described herein. For example, features of methods may be implemented in suitably configured hardware, and the configuration of the specific hardware described herein may be employed in methods implemented using other h a rdwa r e. It will be appreciated from the discussion above that the embodiments shown in the Figures are merely exemplary, and include features which may be generalised, removed or replaced as described herein and as set out in the claims. With reference to the drawings in general, it will be appreciated that schematic functional block diagrams are used to indicate functionality of systems and apparatus described herein. It will be appreciated however that the functionality need not be divided in this way, and should not be taken to imply any particular structure of hardware other than that described and claimed below. The function of one or more of the elements shown in the drawings may be further subdivided, and / or distributed throughout apparatus of the disclosure. In some embodiments the function of one or more elements shown in the drawings may be integrated into a single functional unit. In some examples the functionality of the controllers and data processor described herein may be provided by a general-purpose processor, which may be configured to perform a method according to any one of those described herein. In some examples it may comprise digital logic, such as field programmable gate arrays, FPGA, application specific integrated circuits, ASIC, a digital signal processor, DSP, or by any other appropriate hardware. In some examples, one or more memory elements can store data and / or program instructions used to implement the operations described herein. Embodiments of the disclosure provide tangible, non-transitory storage media comprising program instructions operable to program a processor to perform any one or more of the methods described and / or claimed herein and / or to provide data processing apparatus as described and / or claimed herein. The controllers and data processors may comprise an analogue control circuit which provides at least a part of this control functionality. An embodiment provides an analogue control circuit configured to perform any one or more of the methods described herein. 5 The above embodiments are to be understood as illustrative examples . Further embodiments are envisaged. It is to be understood that any feature described in relation to any one embodiment may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the 10 embodiments, or any combination of any other of the embodiments. Furthermore, equivalents and modifications not described above may also be employed without departing from the scope of the invention, which is defined in the accompanying claims.

Claims

1. A method of biophysical characterisation of a lipid nanoparticle analyte, said method comprising the steps of:i) obtaining wave data for a sample based on physical 5 measurements of a wave generated, in a liquid system, by providing contact between a droplet of the sample and the liquid system, wherein the wave comprises modes encoding characteristics of the interaction between the sample and the liquid system,wherein the sample comprises the lipid nanoparticle analyte, and 10 wherein the lipid nanoparticle analyte comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation; andii) providing a biophysical characterisation of the lipid nanoparticle analyte based on the wave data.

152. A method of identifying a lipid nanoparticle analyte with comparable biophysical properties to a target lipid nanoparticle analyte, said method comprising the steps of:(i) obtaining wave data for a sample based on physical 20 measurements of a wave generated, in a liquid system, by providing contact between a droplet of the sample and the liquid system, wherein the wave comprises modes encoding characteristics of the interaction between the sample and the liquid system,wherein the sample comprises the lipid nanoparticle analyte, and 2 5 wherein the lipid nanopar tide analyte comp rises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation, or is a constituent biomolecule of a lipid nanoparticle formulation, for 3 11 Cl ± V S -L S r(ii) providing a comparison of the wave data obtained from 30 interaction of the sample with the surface of the liquid system, with wave data associated with the target lipid nanoparticle analyte;optionally repeating steps i) to ii) for a different lipid nanoparticle analyte, until a lipid nanoparticle analyte is identified having comparable values of at least one biophysical characteristic,indicative of comparable biophysical properties of lipid nanoparticle analyte and the target lipid nanoparticle analyte.

3. The method of Claim 2, wherein the comparison is based on a biophysical characterisation of the lipid nanoparticle analyte based on the wave data and a corresponding characterisation of the target 1i p i d n a n o p a r t i c 1 e a n a 1 y t e .

4. The method of any preceding claim, wherein the sample comprises a precursor lipid nanoparticle formulation dissolved in an organic solvent.

5. A method of biophysical characterisation of a biomolecule or composition thereof, said method comprising the steps of:i) obtaining wave data for a sample based on physical measurements of a series of waves, each wave of the series being generated, in a liquid system., by providing contact between a series of droplets of the sample and the liquid system, wherein the waves comprise modes encoding characteristics of the interaction between the sample and the liquid system;'wherein the sample comprises the biomolecule, for analysis; and ii) providing a biophysical characterisation of the biomolecule, or composition thereof, based on the wave data, wherein providing contact between the series of droplets of the sample and the liquid system generates a thin film comprising the biomolecule, or composition thereof, on the liquid system, and wherein at least one of the droplets of the series of droplets is contacted with the thin film.

6. The method of Claim 5, wherein the sample comprises the lipid nanoparticle analyte and an organic solvent, and wherein the lipid nanoparticle analyte comprises a lipid nanoparticle formulation or a precursor lipid nanoparticle formulation.7, The method of Claim 5 or Claim 6, wherein series of waves is generated by providing corresponding droplet of the series of dropletsthe wave data of the contact between the and the thin film.

8. The method of Claims 5 to 7, wherein the thin film is generated solely from the lipid nanoparticle analyte.

9. The method of Claims 5 to 8, wherein the biophysical characterisation is further based on the wave data indicating the build-up of the analyte on the liquid system.

10. The method of any preceding claim, wherein the biophysical characterisation comprises a plurality of data values suitable for providing a phase diagram or a polydispersity index for the lipid nanoparticle analyte.

11. The method of any preceding claim, wherein:the lipid nanoparticle analyte is not, and preferably does not comprise, a glyceride oil; and / orthe lipid nanoparticle precursor formulation comprises an amphiphilic lipid.

12. The method of any preceding claim, wherein the constituent biomolecule of a lipid nanoparticle formulation is selected from: cholesterol or a derivative thereof; a cationic and / or ionisable lipid; a phospholipid; a PEGylated lipid; or a surfactant, and / or wherein the lipid nanoparticle formulation comprises an assembled lipid nanoparticle; and / orwherein the lipid nanoparticle formulation or the precursor lipid nanoparticle formulation comprises t’wo or more of a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid; a phospholipid; and a surfactant, preferably wherein the precursor lipid nanoparticle formulation comprises at least a PEGylated lipid and / or a cationic and / or ionisable lipid.

13. The method of any preceding claim, wherein the sample comprises a lipid nanoparticle analyte which comprises a precursor lipid nanoparticle formulation, the precursor lipid nanoparticle composition comprising: a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid; and a phospholipid,preferably wherein the sample; consists of the lipid nanoparticle analyte and a solvent system, wherein the candidate composition consists of a structural lipid; a PEGylated lipid; a cationic and / or ionisable lipid; and a phospholipid.

14. The method of Claim 11 or Claim 12, wherein:the cationic and / or ionisable lipid is an ionisable lipid which comprises a tertiary amine moiety; and / orthe phospholipid is a glycerophospholipid or a phosphosphingolipid, preferably wherein the glycerophospholipid is selected from: phosphatidylcholines such asd i s t e a r o y 1 p h o s p h a t i d y 1 c h o 1 i n e (D S P C) , 1 - p a 1 m i t o y 1 - 2 -o 1 e o y 1 p h o s p h a t i d y 1 c h o 1 i n e ( P 0 P C) , d i p a 1 mi t o y 1 p h o s p h a t i d y 1 c h o 1 i n e (DPPC), 1,2-dipentadecanoyl-sn-glycero-3-phosphocholine anddimyristoylphosphatidylcholine (DMPC); phosphatidylethanolamines such as distearoylphosphatidylethanolamine (DSPE),d i m y r i s t o y 1 p h o s p h a t i d y 1 e t h a n o 1 a mi. n e (DM PE) , a n ddipalmitoylphosphatidylethanolamine (DPPE); phosphatidylserines; p h o s p h a t i d y 1 i n o s i t o 1. s; p h o s p h a t i d y 1 g 1 y c e r o 1 s; a n d p h o s p h a t i d i c acids, and wherein the phosphospingolipid is selected from ceramide pho s ph o r y1cho1i ne, ceramide p hos ph o ry1e th a no1ami ne, a nd ceramide p h o s p h o r y 11 i p i d; a n d / o rthe PEGylated lipid is selected from: PEG-modifiedphosphatidylethanolamines such as DSPE-PEG; PEG-modified phosphatidic acids; PEG-modified ceramides; PEG-modified dialkylamines; PEG-mo d i f i e d d i a c y 1 g 1 y c e r o 1 s ; a n d PEG- mo di f i e d d i a 1 k y 1 g 1 y c e r o 1 s ; a n d / or'the structural lipid is selected from: sterols such as cholesterol; fecosterol; sisosterol; campesterol; stigmasterol; brassicasterol; ergosterol; desmosterol; zymosterol; lanosterol.

15. The method of any preceding claim, wherein the sample comprises a lipid, nanoparticle analyte which is a lipid nanoparticle formulation or a. precursor lipid nanoparticle formulation comprising:a cationic and / or ionisable lipid in an amount in the range of from 30 to 70 mol%, preferably from 40 to 60 mol%, and more preferably from 45 to 55 mol% of the lipid nanoparticle analyte; and / ora phospholipid in an amount in the range of from. 0.1 to 20 mol%, preferably from 0.5 to 15 mol%, and. more preferably from 1 to 12.5 mol% of the lipid nanoparticle analyte; and / ora structural lipid in an amount in the range of from 20 to 60 mol%, preferably from 30 to 55 mol%, and more preferably from 35 to 50 mol% of the lipid nanoparticle analyte; and / ora PEGylated lipid in an amount in the range of from 0.1 to 20 moll, preferably from 0.5 to 15 mol%, and more preferably from 1 to 12.5 mol% of the lipid nanoparticle analyte.

16. The method of any preceding claim, wherein the biophysical characterisation comprises one or more properties selected from: compressibility, saturation pressure, surface activity, viscoelasticity, viscosity, stability in a particular environment, hydrophobicity, binding affinity to a particular target, solubility in a. particular solvent, surface charge, specific gravity, and phasetransition .

17. The method of any preceding claim, wherein the method further comprises a step iii) of providing a metric of developability of the lipid nanoparticle analyte based on the biophysical characterisation.

18. The method of Claim 16, wherein the metric of developability is selected from one or more of: in vivo performance such as biodistribution, circulation time, reduction in mortality or morbidity, and protein expression level; in vitro performance such as transfection efficacy; encapsulation efficiency; polydispersity index; lipid nanoparticle size; and stability.

19. The method of any preceding claim, wherein the wave data comprises a representation of a plurality of said waves generated in the liquid system, each of those waves corresponding to contact of a different one of said droplets with the liquid system.

20. The method of any preceding claim, wherein providing a biophysical characterisation comprises identifying a plurality of separable contributions to the variance of the wave data.

21. The method of Claim 20, comprising:obtaining association data defining a relation between at least one of the separable contributions and a biophysical characteristic, anddetermining data indicating the biophysical characteristic, based on the separable contributions and the association data, wherein the biophysical characterisation of the candidate compound or composition provided at step (iii) comprises the data indicating the biophysical characteristic.

22. The method of20or 21,comprising applying adimensionality reduction method to the wave data to identify theseparable c o n tributions.

23. The method of Claim 22, wherein the dimensionality reduction method comprises a blind signal separation, BSS, method.

24. The: method of Claim 23, wherein the BSS method comprises at least one method selected from the list comprising: principal component analysis; singular value decomposition; independent component analysis; and non-negative mat r i x factorization.The method of Claim 24, wherein the dimensionality reductionmethod comprises a multi-resolution analysis method.

26. The method of Claim 25, wherein the multi-resolution analysis method comprises at least one method selected from the list comprising :a time-frequency analysis, such as analysis based on a short-time fourier transform;a wavelet analysis, such as analysis based on a short-time fourier trans form.

27. The method of any preceding claim, further comprising performing the physical measurements of the wave.

28. The method of any preceding claim, further comprising providing the contact between the droplet of the sample and the liquid system to generate the wave.

29. The method of any preceding claim., further comprising preparing the sample for analysis.

30. The method of any preceding claim, wherein the wave comprises a s u r f a c e wa ve .

31. The method of Claim. 30, wherein the modes encoding characteristics of the interaction between the sample and the liquid system comprises a Lucassen wave mode and the physical measurement comprises a sensitivity to said Lucassen w^ave mode.

32. A process for manufacturing a product comprising a lipid nanoparticle composition or a precursor lipid nanoparticle composition, the process comprising:obtaining wrave data for a sample based on physical measurements of a wrave generated, in a liquid system, by providing contact between a droplet of the sample and the liquid, system, wherein the sample comprises a lipid nanoparticle formulation, a precursor lipid nanoparticle formulation, or a. constituent biomolecule of a lipid nanoparticle formulation and wherein the wavecomprises modes encoding characteristics of the interaction between the sample and the liquid system;providing a biophysical characterisation of the lipid nanoparticle composition or precursor lipid nanoparticle composition 5 based on the wave data; andmanufacturing the product.

33. The process of Claim 32, wherein the biophysical characterisation is provided by the method of any of Claims 1 to 30. 1015s

Images