DE112019000371B4 - Method for screening psychoactive substances using an NMR spectrometer having a permanent magnet - Google Patents

Abstract

A method of screening a sample comprising a test compound for psychoactive substances, the method comprising providing a benchtop permanent magnet NMR spectrometer having a magnetic field strength of 0.5 Tesla to 2.5 Tesla and executing a computer program to perform subsequent steps :(a) obtaining an NMR spectrum of the sample using the benchtop NMR spectrometer, and generating a measurement spectrum set using the NMR spectrum, the measurement spectrum set consisting of one or more shifted spectra, each shifted spectrum associated with an NMR spectrum at a different frequency shift;(b) comparing the measurement spectrum set to one or more reference spectra, each reference spectrum relating to a psychoactive substance and recorded using a magnetic field strength of 0.5 Tesla to 2.5 Tesla; and (c) categorizing the test compound into a psychoactive substance class based on the comparison.

Description

FIELD OF THE INVENTION

The invention relates to a method for screening psychoactive substances using an NMR spectrometer which has a permanent magnet with a magnetic field strength of 0.5 to 2.5 Tesla.

BACKGROUND OF THE INVENTION

There is a need to quickly and accurately screen suspected contraband confiscated by law enforcement agencies to determine the presence of psychoactive substances contained therein. A situation of particular need is in prisons, which have seen a dramatic increase in the use of 'new psychoactive substances' (NPS) in recent years. Controlled drugs are currently detected in UK prisons either by technical analysis of biological samples (such as urine, hair and blood) or by ion mobility scanners. For established substances, these methods are generally very effective, however, these methods have strong limitations in terms of selectivity (a potential for false positives), sensitivity (especially when a small amount of psychoactive substances are present in a mixture), an ability new or previously identify unreported psychoactive substances and the processing time between sample collection and identification.

Raman spectroscopy has recently been considered for NPS detection, and while portable systems are commercially available, they are generally limited by low sensitivity and the inability to detect unknown psychoactive substances in bulk mixtures.

It is desirable to provide a screening process for psychoactive substances that is easy to implement, such that a result can be obtained by a security officer who has seized the suspected intoxicant without the need for a trained chemist or extraction biological samples. Moreover, this process should be realized using a small, inexpensive device, so that its adoption in prisons, police stations, airports, and the like is favored.

The present invention is directed to solving these problems.

Nuclear magnetic resonance (NMR) spectroscopy provides a method for studying the electronic environment surrounding NMR-active nuclei and derived structures of the molecules to which these nuclei are bound. NMR spectroscopy involves placing a sample containing NMR active nuclei (such as hydrogen-1) in a static external magnetic field Bo generated by an NMR spectrometer. A radio frequency (RF) electromagnetic pulse is then applied to the sample by the spectrometer at a frequency characteristic of the NMR active nucleus of interest. The energy provided by this pulse is "absorbed" by these nuclei and re-radiated as radio waves at the resonant frequency of the active nucleus. The signal is known as the free induction decay signal (FID) and is detected and analyzed by the spectrometer to deduce the chemical structure of this sample.

High-field NMR spectrometers are routinely used by chemists, for example when synthesizing new compounds, to verify the structure of the compound produced or to fully elucidate the structure of an unknown compound. These spectrometers typically require large and expensive superconducting magnetic assemblies and skill to operate. In addition, it often takes several days to identify the structure of a connection using such devices. This is due to the number of different NMR experiments that must be performed to generate sufficient data to fully elucidate the structure.

Low-field NMR spectrometers have recently been developed as a cheaper, more compact, and portable device to perform certain types of NMR experiments that were otherwise performed in high-field NMR spectrometers. Rather than requiring dedicated infrastructure, trained personnel, and extensive installations, benchtop NMR spectrometers use permanent magnets that generate a magnetic field between 0.5 and 2.5 Tesla at the spectrometer's sample holder. In some cases, this field strength may be too small to clarify the chemical structure of an unknown compound.

Benchtop NMR spectrometers are generally self-contained units that can be placed directly on a laboratory bench or surface and moved around as needed. These spectrometers are usually straightforward to use by a non-specialist user. An example One such benchtop NMR spectrometer is the Pulsar™ provided by the Oxford Instruments Group. Although these devices have a wide potential for use, their use has hitherto been limited to chemists, for example in the food and beverage industry, and for educational purposes.

DE 101 19 455 A1 discloses a method for evaluating data that is generated using a magnetic resonance technique and contains spectroscopic information, with at least two peaks of a spectrum of the data being detected and with the detected peaks having a position relative to one another that the detected peaks have within the spectrum Peaks of known substances are compared and known substances are assigned to the detected peaks in the event of a match of the comparing.

SUMMARY OF THE INVENTION

1. (a) Erhalten eines NMR-Spektrums der Probe unter Verwendung des NMR-Spektrometers und Erzeugen eines Messspektrumsatzes unter Verwendung des NMR-Spektrums, wobei der Messspektrumsatz aus einem oder mehreren verschobenen Spektren besteht, wobei jedes verschobene Spektrum mit einem NMR-Spektrum bei einer unterschiedlichen Frequenzverschiebung korrespondiert;
2. (b) Vergleichen des Messspektrumsatzes mit einem oder mehreren Referenzspektren, wobei sich jedes der Referenzspektren auf eine psychoaktive Substanz bezieht und mittels einer magnetischen Feldstärke von 0,5 bis 2,5 Tesla erfasst wurde; und
3. (c) Kategorisieren der Testverbindung in eine Klasse psychoaktiver Substanzen basierend auf dem Vergleich.

A first aspect of the invention provides a method for screening a sample comprising a test compound for psychoactive substances using an NMR spectrometer having a permanent magnet with a magnetic field strength of 0.5 to 2.5 Tesla, the method comprising the execution of a computer program for carrying out the following steps:

1. (a) Obtaining an NMR spectrum of the sample using the NMR spectrometer and generating a measurement spectrum set using the NMR spectrum, the measurement spectrum set consisting of one or more shifted spectra, each shifted spectrum having an NMR spectrum at a different frequency shift corresponds;
2. (b) comparing the measurement spectrum set with one or more reference spectra, each of the reference spectra relating to a psychoactive substance and being acquired using a magnetic field strength of 0.5 to 2.5 Tesla; and
3. (c) categorizing the test compound into a psychoactive substance class based on the comparison.

A resulting advantage of this is that a test compound can be screened for the presence of psychoactive substances and any psychoactive substances found can then be categorized in a quick and user-friendly manner. The results can then be displayed to a user, for example by means of a computer monitor. The comparison performed in step (b) represents a differentiator with respect to traditional integration-based NMR methods, which generally depend on higher resolution data and manual interpretation. The method can then be implemented (and is implemented) using a low field NMR spectrometer with a permanent magnet. Such devices are commonly referred to as "benchtop" NMR spectrometers and have associated advantages of ease of use, affordability, size and weight. These devices are also sufficiently sensitive to detect small amounts of psychoactive substances (e.g. less than 10 mg) or several different psychoactive substances in a sample in a short period of time (e.g. less than 10 minutes).

There has been a general bias among chemists against using low-field NMR spectrometers for compound screening. This is related

Together, keep in mind that the field strength may not be high enough to obtain unambiguous results regarding the chemical structure of the compound, especially if the sample contains a mixture of different compounds. For example, it is normal for there to be significant spectral overlap in NMR data acquired at low magnetic field strengths, which can make it difficult for a user to identify a compound according to its spectral characteristics. However, we have found that a low-field NMR spectrometer can be used to allow identification of the psychoactive substance class, even if it is not always possible to identify its exact chemical structure with a sufficiently high level of confidence. Further, while the presence of spectral overlap can make manual interpretation of the data difficult, it generally does not affect the ability to categorize or identify the test compound once the measurement spectrum set (also referred to herein simply as the "measurement spectrum") is compared to one or more reference spectra . The ability to quickly on-site assess the presence of a controlled substance, as well as its class, can be extremely useful to a security or law enforcement officer. If further testing is needed to identify the chemical composition of the psychoactive substance, the sample can be sent to a laboratory for a full analysis.

The use of benchtop NMR spectrometers to screen for psychoactive substances (sometimes referred to as "controlled substances" net) is particularly desirable in the security and law enforcement arena, as no specialized training is required to operate these devices. Furthermore, since the comparison is performed using a computer, rather than requiring a manual assessment to be performed, the screening process is made overall easier for the layperson. This is in contrast to traditional screening processes, which include those using high-field NMR spectrometers.

In particular, in the context of drug screening, NMR technology offers the advantage that the class of psychoactive substances can be identified even if this substance is newly synthesized and has not been identified in the past. The likely effects of the substance, particularly any psychoactive effects, can therefore be predicted according to the class of psychoactive substance to which it belongs. Appropriate action can then be taken if a person is suspected to have ingested that substance

The measurement spectrum is compared to one or more reference spectra recorded at a magnetic field strength of 0.5 to 2.5 Tesla. The one or more reference spectra were preferably recorded at the same field strength as the measurement spectrum. Furthermore, these spectra are preferably acquired using a permanent magnet NMR spectrometer. Typically, the one or more reference spectra are pre-acquired at a remote location. However, the spectrometer is preferably of the same model or type as that used in step (a). The reference spectra acquired using the same test compound analyzed in step (a) may therefore have a similar spectral response to the measurement spectrum, aiding in the comparison process performed. In contrast, if the measurement spectrum is compared to reference spectra acquired using a high-field NMR spectrometer, manual intervention (e.g., a trained chemist) may be required to perform step (c).

In practice, the measurement spectrum can comprise a processed NMR spectrum or a processed set of NMR spectra. For example, the measurement spectrum may include a normalized form of an NMR spectrum obtained for the sample. Step (a) involves acquiring an NMR spectrum of the sample using the spectrometer and generating one or more shifted spectra from the NMR spectrum, each shifted spectrum corresponding to the NMR spectrum on a shifted frequency scale. The measurement spectrum therefore contains the one or more shifted spectra. The one or more shifted spectra can thus each take the place of the measurement spectrum in the comparison in step (b). Changes in test compound concentration can cause a frequency shift in the resulting NMR spectrum. Although the concentration of the substance(s) used to form the one or more reference spectra may be known, the concentration of the test compound is unknown. Forming a set of "shifted spectra" therefore increases the likelihood that one of the sample spectra will match a reference spectrum, aiding the comparison process in step (b).

Step (b) preferably has the following steps: (i) Carrying out a similarity comparison between the measurement spectrum and the one or more reference spectra. For example, in step (i) a similarity comparison can be performed between each shifted spectrum and the one or more reference spectra. Each of the reference spectra in step (i) is preferably a compound reference spectrum that corresponds to an NMR spectrum obtained from a particular psychoactive compound. Alternatively, the respective reference spectrum in step (i) can also relate to a class reference spectrum which has an average spectral response for compounds within a specific class of psychoactive substances.

It is desirable to test mixtures of psychoactive substances as part of the screening process. The similarity comparison in step (i) is therefore preferably carried out between the measurement spectrum and a plurality of reference spectra, and step (b) preferably additionally has the following steps: (ii) selecting one or more combinations of the reference spectra based on the in step (i ) executed similarity comparison; (iii) calculating one or more mixture reference spectra, each mixture reference spectrum corresponding to an addition of a respective combination of reference spectra selected in step (ii); and (iv) performing a similarity comparison between the measurement spectrum and the one or more mixture reference spectra. The mixture reference spectra are therefore calculated instead of being measured empirically. This reduces the number of reference spectra that need to be stored in memory and the number of experiments required to implement the screening process.

In order to limit the processing time required, the combinations selected in step (ii) may correspond to a subset of the reference spectra compared in step (i), such as those found to have the highest similarity (corresponding to a lowest correlation distance) with have the measurement spectrum. The comparison process can therefore be adaptive, with the output of one step affecting the actions taken in the next step, and so on. Certain combinations of reference spectra may be prohibited in step (ii) based on the realization that the combination of the corresponding substances does not occur in practice (e.g. synthetic cannabinoids with steroids). This further reduces the computing load. In the event that the reference spectra in step (i) are compound reference spectra, the one or more mixture reference spectra correspond to mixtures of the corresponding compounds. Alternatively, if the reference spectra in step (i) are class reference spectra, the mixture reference spectra in step (ii) correspond to mixtures of these classes.

It is particularly desirable to consider different combinations and weights of mixtures as part of the comparison. For example, the method may additionally comprise iterating steps (ii) to (iv) using different combinations of selected reference spectra to form the mixture reference spectra in each iteration until a matching spectrum with a sufficiently high similarity is found, or until a threshold number of iterations is performed. The different combinations can have different weightings of the reference spectra. In the above iterative process, each iteration can be performed without reference to the last. While this is quick to implement and execute, more accurate results can be obtained if the process is adaptive, for example such that it approaches a solution. The similarity comparison in step (iv) is therefore preferably carried out between the measurement spectrum and a plurality of the mixture reference spectra, and step (b) further comprises the following steps: (v) selecting one or more of the mixture reference spectra based on the similarity comparison in step (iv) ; (vi) adjusting the ratios of the reference spectra that form the mixture reference spectra to form one or more adjusted mixture reference spectra; and (vii) performing a similarity comparison between the measurement spectrum and the one or more adjusted mixture reference spectra. Again, it is desirable to reduce processing time by limiting the number of mixture reference spectra selected. For example, only a subset of the mixture reference spectra calculated in step (iii) can be selected in step (v), such as those that have the greatest similarity to the measurement spectrum. Adjusting the ratios in step (vi) then produces still further reference spectra for comparison with the measurement spectrum, increasing the likelihood of identifying a matching spectrum with a sufficiently high confidence score. Steps (v) to (vii) could also be an iterative process, repeated until a matching spectrum with a sufficiently high similarity is found, or until a threshold number of iterations are performed. Alternatively, step (iv) may comprise performing a non-negative least squares spectral fit of the measurement spectrum to a subset of matching mixture reference spectra obtained from step (iv). This process is typically more computationally demanding (especially over large data sets) but has been found to be more accurate in estimating the correct ratios and components of any psychoactive compounds in the sample at hand.

The more reference spectra that are combined or their weights adjusted, the greater the likelihood that false positives will be returned (where the comparison process incorrectly identifies the measurement spectrum as corresponding to a particular class or compound of psychoactive substances). Therefore, it is desirable to apply a penalty factor to the matched mixture reference prior to any comparison. Thus, step (iv) may further comprise applying a penalty factor to the matched mixture reference spectra. Likewise, step (vii) may additionally include applying a penalty factor to the matched adjusted mixture reference spectra.

It is further desirable that step (b) further includes a process for generating a list of the psychoactive substances corresponding to the spectra compared in each of the similarity comparisons to the measurement spectrum, the psychoactive substances being classified according to the similarity of the corresponding reference spectra can be ranked with the measurement spectrum based on the similarity comparison. This process can constitute step (viii). Step (c) may then involve categorizing the test compound into a psychoactive substance class based on the psychoactive substance identified in the list as corresponding to the best spectrum match. For the For example, a correlation or confidence score may be assigned to each of the spectra or the corresponding substances that are compared to the measurement spectrum (including any shifted spectrum, if any). Step (c) may then include a process of categorizing the test compound into a psychoactive substance class based on the psychoactive substance identified in step (viii) as corresponding to the best matching spectrum.

Numerous different methods can be used for any of the above similarity comparisons. For example, similarity matching can be performed using a distance metric such as a city block metric. Other potential metrics that can be used include Euclidean distance, cosine distance, and correlation distance. The correlation distance (defined as 1 minus the Pearson correlation coefficient between measurement vectors) between the measurement spectrum and each of the reference spectra can be taken and a closest match (i.e., greatest similarity) can be identified with the one that has the smallest correlation distance. These distances can be calculated using the entire spectral range or based on smaller, selected spectral ranges.

The similarity comparison can be calculated based on a tolerance parameter, preferably with a tolerance value of ±0.1 ppm. The tolerance parameter defines the range over which the spectra are "shifted" and thus provides an extent of latitude of the precise positions of the spectral features. NMR spectra can undergo a frequency shift at all resonances, typically due to sample concentration effects. The x-axis of an NMR spectrum is given in parts per million (ppm), which is calculated as the frequency of a resonance divided by the fundamental resonance frequency of the magnetic field. The frequency shift is almost negligible for high-field NMR spectrometers, where the fundamental resonance frequency is high (e.g., greater than 400 MHz), but this shift becomes more pronounced for NMR spectra generated by low-field NMR spectrometers perceptible. This frequency shift is therefore preferably corrected for automatic use of the tolerance parameter during data analysis.

The comparison in step (b) preferably includes a comparison of the measurement spectrum, which has absolute value data, and the one or more reference spectra. Optionally, the one or more reference spectra also have magnitude data for this comparison. For this reason, the measurement spectrum does not have to be phase-corrected. Phase matching typically requires manual intervention and therefore the use of non-phase corrected data means that steps (a)-(c) can be automated more easily. The screening process can thus be completed in less time than when the comparison is performed based on phase-corrected absorbance spectra. Magnitude spectral data is often avoided by chemists due to its tendency to degrade the overlap between spectral peaks and increase in base peaks. Furthermore, spectral peaks cannot be integrated using magnitude spectral data: making traditional spectroscopic analysis difficult. However, step (b) supersedes the need for such analytical methods such that these effects no longer affect the ability to identify or categorize the test compound.

The categorization is performed by comparison to one or more reference spectra. The one or more reference spectra may include one or more class reference spectra, each of the class reference spectra being an average of the plurality of NMR spectra each relating to different psychoactive substance compounds within the same psychoactive substance class. A closest match can therefore be found, for example, between the measured spectrum and the class reference spectrum of each class of psychoactive substances. Alternatively or additionally, the one or more reference spectra may include at least one compound reference spectrum that corresponds to an NMR spectrum obtained from a psychoactive substance of a particular compound. Compound reference spectra can be obtained for the compounds of psychoactive substances likely to be contained in the present sample. Reference spectra can also be obtained from non-psychoactive substances (typically those which are commonly combined with psychoactive substances) for comparison with the measurement spectrum.

The sample can contain a number of different substances, each of which is represented by the measurement spectrum. However, the signal resulting from a psychoactive substance present in the sample is referred to herein as the "psychoactive fingerprint". Step (b) typically involves identifying a fingerprint of a psychoactive substance in the measurement spectrum, the fingerprint of the psychoactive substance ven substance corresponds to the reference spectrum. This can be the case, for example, when a "match" is found between a component of the measurement spectrum and a reference spectrum.

A particular advantage is where step (c) further comprises assessing the effects of the psychoactive substance when administered by a user. This process can be performed automatically once the test compound has been classified by reference to an electronic library or database. This information can provide the user with useful information that will allow him or her to take appropriate action if a person is suspected of having ingested the substance.

Step (c) may further include identifying the test compound based on the comparison. In particular, the name of the test compound can be identified. This can be possible, for example, if the one or more reference spectra contain designated compound reference spectra and a match was found between one of these spectra and the measurement spectrum. The test compound can then be identified as the matched designated compound.

The sample may contain a variety of psychoactive substances. In this case, step (c) preferably includes a categorization of the respective psychoactive substance into a class of psychoactive substances. Step (c) may also include a process for identifying the relative amounts of the different psychoactive substances in the sample at hand.

Contrary to some existing drug screening methods, the sample can be obtained without the need for a biological sample. The suspected intoxicant itself (i.e., the test compound) may instead be dissolved in a solvent such as deuterated dimethyl sulfoxide. This process is often easier to handle practically. The measurement spectrum can then be aligned with reference to a peak occurring at a known frequency, where the peak is due to the presence of the solvent in the sample. Other solvents can also be used, such as: acetone-d6; chloroform-d; water-d2; N,N-dimethylformamide-d7; dimethyl sulfoxide-d6; methanol-d4; and trifluoroacetic acid-d. Typically, the same solvent is used to form the measurement spectrum and the one or more reference spectra.

A particular advantage resulting from this method is that previously unidentified psychoactive substances can be detected and classified. For example, the test compound can be a compound that differs from all psychoactive substances used to obtain the one or more reference spectra.

The NMR spectrometer can typically be used from 270 to 300 Kelvin. This temperature is particularly related to the temperature of the magnet contained therein and is generally not possible for superconducting magnets. The need for large and expensive cooling assemblies can therefore be avoided. An NMR spectrometer with a permanent magnet with a magnetic field strength of 1.0 to 2.5 Tesla is preferably used.

Steps (a) to (c) can be carried out by a computer which forms part of the NMR spectrometer or which is instead simply in communication with it. These steps can also be implemented automatically. The class of psychoactive substances to which the test compound relates can thus be communicated to the user in a user-friendly manner without a need for special knowledge, for example with regard to chemometrics or NMR.

A particular advantage provided by this method is that the process can be carried out in significantly less time than existing screening methods. For example, steps (a) to (c) can be performed in less than one hour, preferably less than 20 minutes, most preferably less than 5 minutes.

The amount of time required to complete steps (a) through (c) may vary in accordance with the number of acquisition scans performed. A suitable number of acquisition scans can thus be selected to balance the need for rapid analysis versus the need for obtaining a measurement spectrum with a suitably high signal-to-noise ratio. Step (a) therefore preferably comprises performing a number of NMR scans of the sample, the number of NMR scans being calculated in accordance with the weight and/or the material used to form the sample. This calculation is typically performed automatically based on information provided by the user.

A range of different NMR active nuclei can be used. For example can the measurement spectrum may be a hydrogen-1, a carbon-13, a nitrogen-15, a fluorine-19 or a phosphorus-31 NMR spectrum. Typically, the measurement spectrum can be a one-dimensional spectrum.

A second aspect of the invention provides a computer program product set up to carry out the method of the first aspect. The second aspect shares advantages similar to those discussed in connection with the first aspect.

character list

• 1 eine perspektivische Ansicht eines Auftisch-NMR-Spektrometers ist, welches zur Verwendung gemäß einer Ausführungsform mit einem Computer verbunden ist;
• 2 ein Flussdiagramm ist, welches ein Verfahren zum Screening einer psychoaktiven Substanz gemäß einer Ausführungsform veranschaulicht;
• 3 ein Messspektrum ist, welches mit einem Klassenreferenzspektrum gemäß einer Ausführungsform verglichen wird;
• 4 ein Messspektrum ist, welches mit einem Verbindungsreferenzspektrum gemäß einer Ausführungsform verglichen wird; und
• 5 ein Flussdiagramm ist, welches Schritte eines Untersuchungsprozesses in Übereinstimmung mit einer Ausführungsform veranschaulicht.

Embodiments of the invention will now be described with reference to the accompanying figures, in which:

• 1 Figure 12 is a perspective view of a benchtop NMR spectrometer connected to a computer for use in accordance with one embodiment;
• 2 Figure 12 is a flow chart illustrating a method for screening a psychoactive substance according to one embodiment;
• 3 Figure 12 is a measurement spectrum that is compared to a class reference spectrum according to one embodiment;
• 4 Figure 12 is a measurement spectrum that is compared to a link reference spectrum according to one embodiment; and
• 5 Figure 12 is a flow chart illustrating steps of an investigation process in accordance with one embodiment.

DETAILED DESCRIPTION

1 Figure 1 is an illustration of a system 1 including a "benchtop" NMR spectrometer 2 connected to a computer 10 for use in accordance with an embodiment of the invention. The spectrometer 2 has a sample holder 5, which is shown as a cylindrical recess in the spectrometer 2, and two NdFeB permanent magnets 7a and 7b, which are operable at room temperature. The magnets 7a and 7b have opposite north and south poles, so that a static primary magnetic field Bo of 0.5 to 2.5 Tesla is provided at the sample holder 5. In use, a sample (not shown) is placed in a sample holder 5 within the primary magnetic field B ₀ . Two magnets are not necessarily required and other magnet geometries such as a cylindrical magnet can be envisaged.

The NMR spectrometer also includes a set of radio frequency (RF) coils 4 operable between 40 to 110 MHz to provide RF magnetic fields perpendicular to the direction of the Bo field, and a set of Bo shim coils ( not shown), which are used to compensate for inhomogeneities within the Bo field. The HF coils 4 can be used to detect HF signals emitted by a sample within the sample holder 5 and to apply the HF fields to the sample. Additional features such as heaters, a vacuum chamber, and vibration dampeners typically found in benchtop NMR spectrometers may also be provided, but these are not illustrated for the sake of clarity.

The Benchtop NMR Spectrometer 2 is a self-contained unit that can be placed directly on a lab bench or surface and moved around as needed (unlike superconducting NMR spectrometers). The spectrometer 2 is connected to a computer 10 which can be used to analyze the results obtained from the spectrometer 2 . Alternatively, this analysis can be performed remotely by a server (not shown) via an internet connection. The computer 10 has a display for the user to view the results of the analysis.

A method for screening psychoactive substances using the system 1 in 1 is provided with reference to the accompanying flow chart in 2 and the spectra of 3 and 4 discussed.

A sample is obtained in step 101 by placing 3 grams of a material containing a test compound suspected of being a psychoactive substance into a vial containing 1 ml of dimethyl sulfoxide (DMSO). The ampoule is shaken and left for five minutes so that the substance is dissolved in the solvent. Any remaining solids, if any, are then removed from the vial or the liquid solvent is transferred to another container, such as a 5 mm NMR sample tube, using a pipette to form the sample.

In step 102 the sample is placed in a sample holder 5 of the table-top NMR spectrometer 2 . The magnets 7a, 7b generate a magnetic field Bo which is highly homogeneous, so that each part of a sample within the sample holder 5 experiences essentially an identical magnetic field strength in a common direction. Once the sample is inside the NMR spectrometer 2 has been stabilized, for example by stabilizing the temperature and applying "trimming" of the magnetic field, an NMR signal is obtained from the calibration sample.

Computer 10 performs the remaining steps of the method. Each successive step can be performed automatically so that the test compound can be classified in a user-friendly way. A computer program product, for example a data storage unit, is provided such that it contains code arranged to cause the computer 10 and the spectrometer 2 to carry out these steps.

A "90 degree" RF pulse is applied to the sample by the RF coils 4 of the spectrometer 2 . This is achieved by using a 50 Hz radio frequency signal that is pre-calibrated to the known magnetic field strength applied to the sample, causing the hydrogen nuclei in the sample to be deflected out of alignment with the applied magnetic field and to precess around the main field direction. This precession decays and the resulting free induction decay (FID) back into alignment causes a release of energy which is detected by the spectrometer RF coils 4 as the FID signal. The FID signal is digitized by the spectrometer 2 and transmitted to the computer 10 for analysis.

It has been found that the signal-to-noise ratio increases with the square root of the number of repetitions of acquisition using the applied RF signal (known as "scans"), at least for under 100 scans. At step 103, an appropriate number of scans is calculated by the computer 10 based on information provided by the user regarding the weight and/or type of material used to form the sample. Typically, more scans are needed when only a small amount of test compound is present in the sample. In the present embodiment, three scenarios are considered. In a first scenario, if the material used to form the first sample is a powder or liquid weighing more than 10 mg, 16 scans are needed (requiring a total of approximately 3.5 minutes to perform). In a second scenario, when the material is a tobacco-like smoking material suspected of being sprayed with a controlled substance, 64 scans are required (requiring a total of approximately 14 minutes to perform). In a third scenario, if the material is a powder or liquid weighing less than 10 mg, 128 scans are needed (requiring approximately 28 minutes). This may be the case, for example, when only trace amounts of the substance are found on a confiscated item such as a container. It should be understood that the analysis is not limited to a specific number of scans and that these examples are indicative only. In step 104, the selected number of scans are performed, the resulting FID signals are detected, processed by software on the computer 10, and then combined together into an aggregated FID signal.

In step 105, the aggregated FID signal is converted into the frequency domain using suitable frequency analysis software, so that a frequency spectrum is obtained, which is available as magnitude data. Appropriate Fast Fourier Transform software can be used to achieve this to reduce the time spent performing the conversion to the frequency domain.

When the time domain FID signal is Fourier transformed, it typically produces "real" and "imaginary" components. Ideally, the real components should contain all the information related to the absorbance spectrum to allow easy identification of components in the sample. In an actual NMR acquisition (in which the precession of the hydrogen nucleus does not start in phase with the experimental frame of reference), a phase correction must be performed to obtain a pure absorption spectrum as the real Fourier transform output. This phase correction is performed by software that adjusts the real and imaginary components of the Fourier transform to different degrees for each sample. However, it is notoriously difficult to obtain consistent phase corrections across large data sets, which affects the accuracy of statistical analysis. Manual intervention is therefore often required. However, in embodiments of the invention, it has been found that the presence of a psychoactive substance in a sample can be identified with a sufficiently high level of confidence without the need for phase-corrected NMR data. As a result, it is desirable that no such phase adjustment be performed subsequent to the Fourier transform. The measurement spectrum presented as magnitude data can be used for analysis instead. The need for manual intervention and time-consuming process steps are therefore avoided.

The frequency spectrum is then subjected to further processing so that the chemical shift of the spectrum is based with tetramethylsilane as the reference point (as is routine in the context of NMR). A "measurement spectrum" is thereby obtained in step 106, with chemical shift data forming the abscissa of the spectrum and intensity forming the ordinate. Chemical shift data provide information about the particular chemical environments in which the hydrogen nuclei reside, such that a particular chemical shift is related to a particular part of a molecule bearing hydrogen atoms. In the event that the test compound contains a psychoactive substance, a psychoactive substance fingerprint emanating from that substance will form part of the measurement spectrum. The method is then continued in step 107a and/or step 107b.

In step 107a, an area of interest within the measurement spectrum is selected for further analysis. In this case, the area of interest is a relatively narrow area in which the psychoactive fingerprint is expected to become visible. This area is referred to as the "class characteristic area". A class characteristic range of 0.5-2 ppm is suitable for identifying compounds in the following classes: ephenidines, diphenidines and cathinones. More than one class characteristic region can be selected to identify the presence of a compound from other classes. For example, a range of 7-8 ppm can be selected to identify the presence of methylenedioxycathinones. Appropriate class characteristic regions can be identified in a simple manner for the detection of certain classes on the basis of the known common chemical structure for substances in this class.

Any peaks not related to the test compound, such as those related to the solvent DMSO, are excluded from further consideration by the selection in step 107a. Then, in step 108a, the class characteristic range is normalized to compensate for the variable concentration of the test compound.

The method then proceeds from step 108a to 109a wherein the class of psychoactive substances to which the test compound relates is estimated. This is done by comparing the normalized class characteristic regions to a plurality of previously acquired class reference spectra stored in memory. Each of the class reference spectra is an average profile within the class characteristic region obtained from a plurality of NMR spectra, each corresponding to different psychoactive substance compounds within the same chemical class. For example, a class reference spectrum can be obtained for each of the following classes of psychoactive substances: diphenidines, cathinones, ephenides, amphetamines, and methylenedioxycathinones. These reference spectra are typically obtained with a magnetic field B0 corresponding to the measurement spectrum (for example by using the same spectrometer 1) using the same solvent and can be available as magnitude or phase-corrected absorption data.

3 provides an example of a normalized class characteristic region of a measurement spectrum (labeled "unknown test sample") next to a class reference spectrum (labeled "average of ephenidines in the reference database"). The normalized portion of the measurement spectrum can be compared to each of the class reference spectra stored in memory and the closest match found. In this embodiment, the shaded area represents the city block distance between the normalized, class-specific area of the measurement spectrum and the average of all known ephenidines in the reference database. The class reference spectrum for ephenidine is identified as the closest match and accordingly the test compound is classified as an ephenidine.

Steps 107b-109b can also potentially be performed concurrently with steps 107a-109a. In step 107b, another region of interest is selected from the measurement spectrum. In this case the range of interest is a relatively broad range in which the psychoactive fingerprint is expected to become visible, for example 0.5-12 ppm. This area is also known as the "full fingerprint area". Any peak values not related to the test compound, such as those related to the solvent DMSO, are excluded from further consideration by this selection. Then, in step 108b, the full fingerprint area is normalized to correct for a variable concentration of the test compound.

The method then continues from step 108b through 109b, identifying the specific compounds for any psychoactive substances present in the sample. This is performed using a similarity match, comparing the normalized complete fingerprint area to a plurality of previously acquired compound reference spectra which are stored in memory. Each of the compound reference spectra relates to a previously measured NMR spectrum from a sample containing a psychoactive substance of a known compound. These reference spectra are acquired at the same basic magnetic field strength and they use the same solvent as the measurement spectrum. These reference spectra can be present as magnitude data or phase-corrected absorption data.

The nearest neighbor between the normalized full fingerprint region of the measurement spectrum and the respective compound reference spectra is determined using a distance metric. In the present invention, the correlation distance is calculated. In particular, the cross-correlation constrained to a certain tolerance range is found for the normalized full fingerprint range of the measurement spectrum with each of the compound reference spectra. This gives correlation distances for the test compound from each compound reference spectrum. The test compound is then identified as the psychoactive substance that relates to the compound reference spectrum with the smallest correlation distance from the normalized full fingerprint range of the measurement spectrum (hence corresponding to the most correlated reference spectrum). Optionally, a threshold can be implemented for the distance above which the test compounds are considered too dissimilar to the reference compounds. Such test connections can be displayed as "unknown" or "unseen before".

4 provides an example of a normalized full fingerprint region of a measurement spectrum (labeled "test compound") next to the most closely correlated compound reference spectrum (labeled "4-fluoroamphetamine reference spectrum"). These two spectra are not exactly aligned along the x-axis due to concentration effects, but this can be corrected by using a tolerance parameter, for example ±0.1 ppm. In this example, the sample was prepared using a test compound related to a different psychoactive substance than that in 3 . Specifically, the test compound is identified as 4-fluoroamphetamine. The class of psychoactive substance to which the test compound relates can then be identified or inferred from this information. The method has been found to be particularly effective in identifying steroids, synthetic cannabinoids and opioids such as fentanyls and heroin.

In the case where the test compound relates to a new, previously unreported psychoactive substance, it may not be possible to identify the compound to which the psychoactive substance relates according to steps 107b-109b. In this case it should still be possible to identify a class of the psychoactive substance present in the sample according to steps 107a-109a.

The results of steps 109a and 109b are displayed to a user in step 110, for example by means of a computer monitor. A confidence rating is typically provided alongside each result, particularly where the name of the test compound is identified. The confidence score may be a correlation distance, in which case a lower number indicates a higher confidence score. Optionally, a plurality of likely compounds and/or classes may be identified and ranked according to their confidence score. In addition, the effects of the test compound ingested by an individual can be reported, enabling the user, who may be a security or law enforcement officer, or a medical practitioner, to take appropriate action. If the presence of several different psychoactive substances is detected within the sample, the respective classes and compound designations (if available) for each of the substances can then be displayed. Optionally, the relative amounts of these substances can also be given.

The compound reference spectra and the class reference spectra can be stored locally within the computer 10 or the spectrometer 2, respectively. However, these are preferably stored in a remote database, which is accessed by the computer 10 via the Internet. This allows the database to be continuously updated with additional reference spectra from a remote location.

A further embodiment of the invention will now be described with reference to the flowchart in 5 described. This embodiment is directed to a process for identifying the exact compound of a psychoactive substance present in the sample (equivalent to steps 107b-109b of Fig 2 ), however, a similar process can be performed to easily identify the class of psychoactive substance in the sample, as will be described later. The method below uses a minimum distance metric to match a measured proton magnitude spectrum of a sample to a database of stored reference spectra same. Although this method is based on a calculation of the correlation between the spectra, other distance metrics can be used.

The procedure runs as in the 2 described steps 101 to 105. The frequency spectrum is then subjected to additional processing by calculating the chemical shift for the spectrum based on tetramethylsilane as a reference point, and the full fingerprint region of the NMR spectrum is selected and then normalized (as before). At this point, additional processing is then applied to the frequency spectrum. Specifically, in step 201, the spectrum is converted into a "shifted" matrix by repeating the spectrum along the chemical shift axis (ie, the x-axis in 4 ) is shifted to simulate the frequency shift effect from a range of sample concentrations. In general, the concentration of a test compound will not match the concentration of the compounds used to generate the reference spectra. The process of creating a shifted matrix produces a set of different spectra that have different axial offsets (corresponding to different possible concentrations), which can then be compared to the reference spectra. This reduces the likelihood of an unsatisfactory match along the chemical shift (ppm) scale, which could otherwise compromise the identification process. These spectra are referred to here as “shifted spectra”.

The method then continues at step 202, at which point the correlation between each of the shifted spectra and each of the plurality of compound reference spectra is calculated. This process is performed using a collection of direct matrix operations. For each of the compound reference spectra, only the highest correlation value for the shifted spectra set is kept. Then, in step 203, a single compound match list is returned which ranks the compound reference spectra (and/or their corresponding compounds) based on a descending order of correlation.

The method then continues from step 203 to step 204, using the ranking returned from step 203 to form "mixture reference spectra" based on pairs of compound reference spectra. The mixture reference spectra are calculated in preference to being measured experimentally and correspond to possible mixtures of the reference compounds. In particular, the spectral components of different compound reference spectra are added, in this case in 1:1 ratios (alternative weights are considered later). The depth of the search is determined by selecting an appropriate number of possible connection combinations, taking into account the available computing power. This number can be varied in accordance with the correlation values output from step 203. For example, if step 203 indicates a strong correlation for certain test compounds, only those test compounds are considered as constituting the mixture. If step 203 indicates only a weak correlation for the different test compounds considered, a higher number of possible combinations of mixtures can be considered. In practice, this can be implemented by only considering those compounds for which the correlation distance was below a threshold. Alternatively, only a fixed number of subsets of compounds are selected in step 204 to form the mixture reference spectra, for example corresponding to the ten lowest correlation distances calculated in step 203. Optionally, certain combinations of compounds to form a mixture reference spectrum (e.g., synthetic cannabinoids with steroids) are not combined, given that these combinations are never encountered in practice. This improves the accuracy of the screening process and reduces processing time.

Next, in step 205, the mixture reference spectra are normalized. For example, a region normalization process can be followed in which these spectra are normalized so that their integrated intensity equals unity. Alternatively, a "proton number normalization" process can be followed (where possible), in which the integrated area of each spectrum is fixed to the number of protons known to contribute to the measurement signal. This helps identify mixtures of compounds with vastly different numbers of protons (e.g. paracetamol has 9 protons and heroin has 23 protons). Alternative possible normalizations can take into account other factors affecting an overall spectral magnitude, such as different compound solubilities.

Correlations of the respective shifted spectra with respective mixture reference spectra are then calculated in step 206 . It fast matrix calculation functions are used to reduce processing time. Cross-mixture correlations (comprising the many "incorrect" matches) are generally slightly higher than for the individual compounds in step 203. In other words, step 206 is more likely to produce false positives than step 202. An "overfitting penalty" is therefore used in Step 207 applied to the correlations from step 206 to compensate for this. The determination of an effective penalty value can be chosen based on the present properties of the database, and adjusted empirically if necessary.

The ratios of the compound reference spectra that make up the mixture reference spectrum are then adjusted in step 208 . In practice, this can be achieved by computing new mixture reference spectra (referred to herein as "adjusted mixture reference spectra") using different weights of the matching compound reference spectra returned from step 203 . Steps 204 through 207 correspond to a first pass to identify any mixtures in the sample. Their efficiency reduces the overall computational power and the time required to estimate the test compound(s) present in the sample. In particular, the output from step 206 can be used to inform which reference spectra to consider in step 208 . For example, mixture reference spectra are formed in step 208 only from combinations of compound reference spectra which, when combined in step 204, produced mixture reference spectra that had a correlation distance to one of the fitted spectra that was below a threshold. In principle it would be possible to move step 203 to step 208, but at the expense of a longer analysis time, since step 208 would then be more computationally intensive.

The ratios of the compound reference spectra that form the mixture reference spectra can be adjusted by performing a constrained (non-negative) least squares spectral fit of the shifted spectra to a subset of pairs of most strongly correlated compound reference spectra identified in step 203. A fast algorithm is used for this fitting (e.g. as described by "Fast algorithm for the solution of large-scale non-negativity-constrained least squares problems", Benthem & Keenan, J. Chemometrics 2005). This method returns regression coefficients which can only assume positive values and which roughly indicate the ratios of the different reference compounds present. The correlation of each shifted spectrum with each of the adjusted composite spectra is calculated. A mixture match list is then returned in step 209 indicating, for each of the adjusted mixture reference spectra generated, the weights of the psychoactive compounds used to form the adjusted mixture reference spectra and its correlation with at least one of the shifted sample spectra.

The method then continues at step 210, at which point the entire ranked identification list is generated. Here the correlation lists of the single compound and mixture matches output in steps 203 and 209 are concatenated and sorted to produce the overall ranked identification list. The best match (with the lowest correlation distance, weighted to include any penalty factors applied) is then returned to the user as the identified test compound(s). In the case where that match is a mixture of psychoactive substances, the relative amounts of those compounds can be estimated based on the weighting of the compound reference spectra used to form the corresponding mixture reference spectrum or adjusted mixture reference spectrum. The class of psychoactive substance in the sample can then be identified as corresponding to the best match class of psychoactive substance.

Steps 201 through 209 can be generalized into three main stages. Stage one (steps 201-203) involves correlating the measurement spectrum with the compound reference spectra, stage two (steps 204-207) involves correlating the measurement spectrum with the mixture reference spectra, and stage three (steps 208 and 209) involves optimizing the ratios in the mixture reference spectra . It has been found that the combination of these three steps is particularly effective in accurately identifying psychoactive compounds present in a sample. Specifically, it has been found that this process can identify the association of any psychoactive substance present in a sample with a success rate of over 90% in less than 10 minutes.

In an alternative embodiment, one or more of the above referenced 5 described steps are completed in an iterative manner. For example, steps 204 and 207 may be repeated up to one predefined number of iterations or until a mixture reference spectrum is generated at which the correlation distance is below a threshold value (indicating a sufficiently high level of similarity between the mixture reference spectrum and the measurement spectrum). Similarly, the weights of the components used to form the mixture reference spectra can be adjusted in an iterative manner: either for a threshold number of iterations or until a reference spectrum is generated for which the correlation distance is below a threshold.

A similar process to that related to above 5 can be performed to easily identify the class of psychoactive substance present in a sample. This process could be followed instead of or in addition to the above process (e.g., where a psychoactive compound was not identified at step 210 with a sufficiently high confidence score). In this case, following step 105, instead of selecting the full fingerprint region of the NMR spectrum, the class characteristic region is selected. This area is then normalized and a shifted matrix is formed (as before). The procedure is then as with reference to 5 described, but class reference spectra are used instead of a compound reference spectrum. All calculated mixture reference spectra also correspond to mixtures of classes and not to composite mixtures. Such a process may be faster in terms of computational power, but its results provide less accuracy. In addition, the process is not suitable for all classes, since some do not provide a characteristic signal. Compounds within these classes are better identified and classified using the method previously described, which focuses on identifying the compound itself.

In yet another embodiment, the method can additionally compare the shifted spectra to one or more reference spectra corresponding to non-psychoactive substances. For example, the process in 5 be expanded to account for mixtures of three components, the third component being a non-psychoactive substance such as a background residual solvent signal. Isolating and, optionally, subtracting any signals that arise from non-psychoactive signals can aid in correctly identifying a psychoactive substance in the sample. However, it may also be necessary to apply overfitting penalties to compensate for the higher probability of false positives.

It is to be welcomed that an improved method for screening for psychoactive substances is provided. Advantageously, the class of psychoactive substance present in the sample can be identified in less time than some existing methods, typically less than an hour. In addition, the method is easy to use and sensitive to small amounts of psychoactive substances in a mixture. Furthermore, the method is suitable for identifying the presence of new or previously unreported psychoactive substances. NPS as well as more traditional psychoactive substances such as heroin, cocaine and steroids can therefore be detected.

Claims (20)

A method of screening a sample comprising a test compound for psychoactive substances, the method comprising providing a benchtop permanent magnet NMR spectrometer having a magnetic field strength of 0.5 Tesla to 2.5 Tesla and executing a computer program to perform subsequent steps : (a) Obtaining an NMR spectrum of the sample using the benchtop NMR spectrometer, and generating a measurement spectrum set using the NMR spectrum, the measurement spectrum set consisting of one or more shifted spectra, each shifted spectrum associated with an NMR spectrum at a different frequency shift; (b) comparing the measurement spectrum set to one or more reference spectra, each reference spectrum relating to a psychoactive substance and recorded using a magnetic field strength of 0.5 Tesla to 2.5 Tesla; and (c) categorizing the test compound into a psychoactive substance class based on the comparison. procedure after claim 1 , wherein step (b) further comprises the following step: (i) performing a similarity comparison between the measurement spectrum and the one or more reference spectra. procedure after claim 2 , wherein the reference spectrum in step (i) is a compound reference spectrum which corresponds to an NMR spectrum obtained from a particular compound. procedure after claim 2 or 3 , wherein the similarity comparison in step (i) is performed between the measurement spectrum set and a plurality of reference spectra, and wherein step (b) further comprises the steps of: (ii) selecting one or more combinations of the reference spectra based on the in step (i ) executed similarity comparison; (iii) calculating one or more mixture reference spectra, each mixture reference spectrum corresponding to an addition of a respective combination of reference spectra selected in step (ii); and (iv) performing a similarity comparison between the measurement spectrum set and the one or more mixture reference spectra. procedure after claim 4 further comprising iterating steps (ii) to (iv) using different combinations of selected reference spectra to form the mixture reference spectra in each iteration until a matching spectrum with a sufficiently high similarity is found or until a threshold number of iterations has been performed. procedure after claim 4 , wherein the similarity comparison in step (iv) is performed between the measurement spectrum set and a plurality of the mixture reference spectra, and wherein step (b) additionally comprises the following steps: (v) selecting one or more of the mixture reference spectra based on the similarity comparison in step (iv) ; (vi) adjusting the ratios of the reference spectra forming the selected mixture reference spectra to form one or more adjusted mixture reference spectra; and (vii) performing a similarity comparison between the measurement spectrum set and the one or more adjusted mixture reference spectra. procedure after claim 6 wherein step (vi) comprises performing a non-negative least squares spectral fit of the measurement spectrum set to a subset of matching mixture reference spectra obtained from step (iv). Procedure according to one of claims 3 until 7 , wherein step (iv) further comprises applying a penalty factor to the matched mixture reference spectra. Procedure according to one of Claims 6 until 8th , wherein step (vii) further comprises applying a penalty factor to the matched adjusted mixture reference spectra. Procedure according to one of Claims 4 until 9 , wherein • step (b) further comprises generating a list of the psychoactive substances corresponding to the spectra that are compared to the measurement spectrum set in each similarity comparison, the psychoactive substances in the list according to the similarity of the corresponding reference spectra to the measurement spectrum set are ranked based on the similarity matching; • and wherein step (c) includes categorizing the test compound into a psychoactive substance class based on the psychoactive substance identified in the list as corresponding to the spectrum with the closest match. Procedure according to one of claims 2 until 10 , where each similarity match is performed using a distance metric such as a city block metric. Procedure according to one of claims 2 until 11 , each similarity comparison being calculated on the basis of a tolerance parameter, in particular with a tolerance value of 0.1 ppm. A method according to any one of the preceding claims, wherein the comparing in step (b) comprises comparing the measurement spectrum set having magnitude data with the one or more reference spectra. procedure after claim 1 or 2 or one of the Claims 4 until 13 wherein at least one of the one or more reference spectra comprises a class reference spectrum corresponding to an average of a plurality of NMR spectra each relating to different psychoactive substance compounds within the same psychoactive substance class. A method according to any one of the preceding claims, wherein at least one of the one or more reference spectra includes a compound reference spectrum which corresponds to an NMR spectrum obtained from a psychoactive substance of a particular compound. The method of any preceding claim, wherein step (b) further includes identifying a psychoactive substance fingerprint in the measurement spectrum set, the psychoactive substance fingerprint corresponding to a reference spectrum. A method according to any one of the preceding claims, wherein step (c) further includes identifying the test compound based on the comparison. A method according to any one of the preceding claims, wherein the sample contains a plurality of psychoactive substances and wherein step (c) comprises categorizing the respective psychoactive substance into a class of psychoactive substances. procedure after Claim 18 wherein step (c) further comprises identifying the relative amounts of the different psychoactive substances in the present sample. Computer program product which is suitable for carrying out the method steps of the method according to one of the preceding claims carried out by the computer program.

Images