Calorimetric method for analysing microbial samples

Abstract

The present document is directed to a calorimetric method for determining if a microbial sample is polymicrobial with the same type or different types of microorganisms or monomicrobial and optionally the taxonomic identity of the microorganism(s) The method utilizes a set of metabolically differentiating inoculation media that allow the metabolic signal from the individual microorganisms present in the microbial sample to be split up so that their type and taxonomic identity can be determined.

Description

[0001] CALORIMETRIC METHOD FOR ANALYSING MICROBIAL SAMPLES

[0002] TECHNICAL FIELD

[0003] The present document is directed to the field of calorimetry and the analysis of microbial samples in order to determine if they are polymicrobial and in that case if they comprise a mixture of different microorganisms of the same type or different types. The present method may further allow for determining the taxonomic identity of the microorganism(s) in the sample.

[0004] BACKGROUND

[0005] Calorimetry is a measurement technology for real-time monitoring and dynamic analysis of chemical, physical and biological processes. Over a period of minutes or days, calorimetry determines the onset, rate, extent, and energetics of such processes for specimens in a closed system measured in W.

[0006] For isothermal calorimetry, the heat release at a constant set temperature is measured. The term ‘micro’ is applied for smaller systems where the heat release is in the micro-watt range.

[0007] For isothermal (micro)calorimetry measurements, the sample to be studied is placed in a sealed container in a constant temperature environment. Changes in heat flow due to chemical and biological processes are specific to the type of system studied, e.g., an isolate of any given bacterial or fungal species under specific metabolic conditions will give rise to a heat flow over time graph termed a thermogram. For monitoring of microbial, such as fungal or bacterial, systems, these specific thermograms can be used to derive the microbial species present and the metabolic viability of the microorganisms present in the sample.

[0008] Calorimetry has been used to detect microbial growth in different types of samples. For example, W02007 / 010379 suggests using calorimetry to detect the presence of microorganisms in samples using media suppressing or promoting growth of microorganisms in a sample by collecting the heat flow signal emitted when the microorganism is incubated in the medium. This document shows that the calorimetric curve in a specific medium may depend upon the microorganism growing in the medium. W02007 / 010379 also suggests kits with different media to detect the presence of different microorganisms in a sample. However, this document only demonstrates that calorimetry can be used to detect the growth of pure cultures of Gram-positive and Gramnegative microorganisms but not that calorimetry can be used to distinguish microorganisms of different types in a microbial sample containing different combinations of Gram-positive bacteria, Gram-negative bacteria and fungi.

[0009] Although a calorimetric thermogram may be used to identify microbial species in monomicrobial samples, polymicrobial samples present a challenge as the thermogram will be a combination of the thermograms of the different microorganisms that are present in the sample. Also, it has been shown that when different microorganisms are present, they affect each other in a way that the thermogram obtained is not simply an aggregation of the individual thermograms of the microorganisms present in the sample (see e.g., Vazquez et al., J Therm Anal Calorim, DOI 10.1007 / sl 0973-015-4606-0, March 2015). Therefore, even if calorimetry has been shown to be useful for studying the interaction between different microorganisms in a sample, analysis of a microbial sample to determine if it is polymicrobial or monomicrobial and taxonomic identification of the microorganisms that are present in it has not been possible.

[0010] SUMMARY

[0011] An object of the present invention is thus to overcome or at least mitigate one or more of the problems described herein.

[0012] The present document is therefore directed to a calorimetric method for the analysis of a microbial sample in order to determine if said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities, wherein said types of microorganisms are selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, said method comprising the steps of: a) providing a microbial sample; b) providing a set of metabolically differentiating inoculation media configured to split up a metabolic profile into the different types of microorganisms present in the microbial sample, wherein said set of metabolically differentiating inoculation media comprises a first metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-positive bacterium, a second metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gramnegative bacterium, and a third metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one fungus; c) inoculating and incubating said microbial sample in said first, second and third metabolically differentiating inoculation medium; d) obtaining calorimetric data from the inoculated and incubated sample of step c), thereby obtaining a metabolic profile from the first, second and / or third metabolically differentiating inoculation medium depending on what types of microorganisms are present in the microbial sample; e) based on the metabolic profiles obtained in step d), determining that said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms if a metabolic profile is obtained in at least two of said first, second and third metabolically differentiating inoculation medium; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if a metabolic profile is obtained in only one of said first, second and third metabolically differentiating inoculation medium.

[0013] The first metabolically differentiating inoculation medium may be selected from the group consisting of salt mannitol broth and CLED broth. The first metabolically differentiating inoculation medium may be supplemented with at least one supplement selected from the group consisting of salts, such as sodium chloride and / or lithium chloride; antibiotics, such as colistin polymyxin B; and / or antimicrobials, such as phenyl ethanol.

[0014] The second metabolically differentiating inoculation medium may be selected from the group consisting of MacConkey broth, Hektoen enteric agar, and Sabouraud broth. The second metabolically differentiating inoculation medium may be supplemented with at least one supplement selected from the group consisting of salts, such as eosin Y; sodium taurocholate; sodium selenite; antibiotics, such as vancomycin, cycloheximide, and / or azinomycin; antimicrobials such as brilliant green; and / or saccharides such as lactose and / or glucose.

[0015] The third metabolically differentiating inoculation medium may be selected from the group consisting of Sabouraud broth, malt extract broth, and yeast extract-peptone dextrose broth. The third metabolically differentiating inoculation medium may be supplemented with at least one supplement selected from the group consisting of gentamycin, chloramphenicol, cephalothin, and cefamandole.

[0016] The calorimetric method may, if the microbial sample is determined to be a microbial sample according to step e) i i) , comprise a further a step of determining that said microbial sample is: i) a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if said metabolic profile(s) match(es) two or more monomicrobial metabolic profiles or matches a combined metabolic profile of a polymicrobial sample with known microbial content; or ii) a monomicrobial sample if the metabolic profile(s) match(es) a monomicrobial metabolic profile, wherein said metabolic profile(s) of said microbial sample are obtained from: a. at least one metabolically differentiating inoculation medium of the same type as the first, second or third metabolically differentiating inoculation medium in which a metabolic profile was obtained according to step e)ii) of any one of the preceding claims; and b. at least one non-selective general purpose inoculation medium, such as tryptic soy broth or Mueller Hinton broth. The the first metabolically differentiating inoculation medium, the second metabolically differentiating inoculation medium and the third metabolically differentiating inoculation medium may be the same as described elsewhere herein. The third metabolically differentiation inoculation medium may also be selected from Czapek Dox broth.

[0017] The calorimetric method of the present document may further comprise a step of determining the family, genus and / or species of said microorganisms by analysing said metabolic profiles.

[0018] The calorimetric method may further comprise a pre-incubation and re-loading step performed before incubating said microbial sample in said metabolically differentiating inoculation media, said pre-incubation and re-loading step comprising inoculating and incubating said microbial sample in at least one inoculation medium, calorimetrically following the metabolic activity of the incubated sample, and, when the R value for the metabolic rate is >0.98 for at least about 20 min or when the metabolic activity of the incubated sample reaches the inflection point or within 2 hours after said inflection point is reached, reloading one or more aliquot(s) of said incubated samplein said metabolically differentiating inoculation media before proceeding with the remaining steps of the method. Two or more aliquots with different total metabolic activity may be reloaded and the aliquots in which the metabolic activity directly after reloading is i) below the detection limit and ii) wherein the metabolic activity in the same aliquot(s) is detected between about 15 min and about 6 hours, preferably between about 15 min and about 3 hours, after said re-loading is used for obtaining said metabolic profiles

[0019] The microbial sample may be a clinical sample from a subject, such as a human or animal subject, an environmental sample, a food or feed sample, and / or a purified microbiological sample. The microbial sample may for example be blood, tissue biopsy, synovial fluid, perinatal fluid, cerebrospinal fluid, pleural fluid, pericardial fluid, sonication fluid from an implant, urine, sputum, stool, and / or saliva. The calorimetric method may be performed without first isolating the microorganisms from the microbial sample.

[0020] The analysis of the metabolic profiles obtained by the calorimetric method may be performed by matching them to metabolic profiles obtained from polymicrobial and / or monomicrobial samples with known microbial content. The metabolic profiles may be analysed using an identification algorithm matching a metabolic profile against a database containing metabolic profiles of known microorganisms. The metabolic profiles may be analysed using a machine learning model trained on monomicrobial and polymicrobial samples.

[0021] The present document is also directed to a computer program comprising computer program code, the computer program code being adapted to, if executed on a processor, implement the step(s) of determining of the calorimetric method of the present document.

[0022] The present document is also directed to a computer program product comprising a computer readable storage medium, the computer readable storage medium having the computer program according to the present document.

[0023] The present document is also directed to a computer-implemented method of determining the type and / or taxonomic identity of one or more microorganisms in a microbial sample, said computer-implemented method comprising the steps of: a) receiving calorimetric data from a calorimeter; b) using the calorimetric data to obtain calorimetric features; c) establishing a correlation between said calorimetric features and a plurality of microorganisms; determining the type and / or taxonomic identity based on said calorimetric features.

[0024] The present document also discloses computer implemented method for training a machine learning model to identify one or more microorganisms, the method comprising: -obtaining calorimetric feature data, the calorimetric feature data being derived from calorimetric data being indicative of the metabolic activity of the one or more microorganism(s) in a set of different metabolically differentiating inoculation media as defined herein;

[0025] -obtaining media data indicative of the set of different metabolically differentiating inoculation media;

[0026] -obtaining microorganism data indicative of a set of microorganisms from which the one or more microorganisms are identified;

[0027] -generating a trained machine learning model by training, using training data comprising the calorimetric feature data, the media data and the microorganism data, a machine learning model to output, based on input operating data, identification of one or more microorganisms.

[0028] The present document is also directed to a kit for performing a calorimetric method, said kit comprising or consisting of a) a set of metabolically differentiating inoculation media; b) a training data set which contains metabolic profiles of known microorganisms incubated in the same metabolically differentiating inoculation media; c) a software component comprising a function that determines the type and / or taxonomic identity of a microorganism.

[0029] The present document is also directed to method for treating a microbial infection in a patient in need thereof, said method comprising determining by the calorimetric method of the present document if said microbial sample is:

[0030] -a polymicrobial sample comprising at least two different types of microorganisms; or -a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities, and, based on the type(s) of microorganisms present in the microbial sample, treating said patient with antimicrobial(s) directed to said type(s) of microorganisms.

[0031] Other features and advantages of the invention will be apparent from the following detailed description, drawings, examples, and from the claims.

[0032] DEFINITIONS AND ABBREVIATIONS

[0033] By “microbial sample” and the like is herein intended a sample potentially comprising microorganisms in the form of bacteria and / or fungi.

[0034] “Taxonomic identity” and the like in the context of the present document refers to the taxonomic identity of the microorganism from the level of family to the level of species.

[0035] By “polymicrobial” is in the context of the present document intended a combination of microorganisms of more than one taxonomic identity. A polymicrobial sample thus comprises a combination of two or more different types of microorganisms (herein also denoted “poly mixtum”) or a combination of two or more different microorganisms of the same type but of different taxonomic identities (herein also denoted “poly similis”).

[0036] By “monomicrobial” is in the context of the present document meant that a microorganism of only one species is present. A monomicrobial sample therefore only has one microbial species in it. A “type” of microorganism is herein defined as a Gram-positive bacterium, a Gramnegative bacterium or a fungus. By “polymicrobial sample comprising at least two different types of microorganisms” is therefore intended a polymicrobial sample comprising any combination of Gram-positive bacteria and Gram-negative bacteria, Gram-positive bacteria and fungi, Gram-negative bacteria and fungi or Gram-positive bacteria, Gram-negative bacteria and fungi. By “polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities” is intended a polymicrobial sample containing two or more microorganisms of the same type, i.e. two or more different Gram-positive bacteria, two or more Gram-negative bacteria or two or more fungi, but where the microorganisms have different taxonomic identities.

[0037] “Calorimetry” and the like refers to a process of measuring the quantity of heat released or absorbed during a chemical reaction, such as during incubation of a microorganism in an inoculation medium. The heat changes in the sample over time [J / s vs. time] may be recorded in a thermogram. Calorimetry according to the present document is performed as known by the skilled person. In short, for the determination of a calorimetric signal from a microbial sample, the sample is inoculated in a vial comprising an inoculation medium and the vial is then placed in the temperature-controlled chamber of the calorimetric device for incubation. The calorimetric signal can be continuously followed and analysed and presented in a so called thermogram displaying the energy changes over time. The calorimetric signal reflects the metabolic activity of the microorganism(s) potentially present in the sample. The calorimetric signal may therefore in the context of the present document also be denoted a metabolic signal, i.e. a signal reflecting the metabolic activity in a sample. The calorimetric data obtained from the thermogram can be used to obtain one or more different calorimetric features (see below) that in turn make up a metabolic profile of the microorganism(s) in the microbial sample.

[0038] By “metabolic profile” is in the context of the present document intended a calorimetric profile over one or more calorimetric features obtained by incubating a microbial sample in a metabolically differentiating inoculation medium. A metabolic profile is a profile of the metabolic activity in a microbial sample. A metabolic profile as referred to herein may be an aggregated metabolic profile of a combination of two or more different microorganisms or it may be a metabolic profile of a monomicrobial component. The present methods of the present document allows for splitting up a metabolic profile into the respective microbial components so that their type and / or taxonomic identity can be determined. A metabolic profile according to the present document may also be represented by the thermogram as such. The metabolic profile may be obtained by following the heat changes in real time by so called real-time calorimetry. The calorimetry used according to the present document may be isothermal (micro)calorimetry. “Metabolic activity” refers to all biochemical reactions that occur in a microorganism, during uptake and utilization of inorganic or organic compounds required for growth and maintenance of a cellular processes. Metabolic activity of microorganisms is in the context of the present document used to obtain a calorimetric metabolic profile (herein usually denoted “metabolic profile”). A metabolic profile may be obtained using one or more different calorimetric features obtained by following the calorimetric signal over time (i.e. heat change over time. “Metabolic rate” is metabolic activity over time, i.e. heat / time, e.g. expressed in J / s, also called heat flow.

[0039] As used in this specification and the appended claims, the singular forms ”a”, ”an” and ’’the” include plural referents unless the context clearly dictates otherwise.

[0040] The term “comprising” includes the term “consisting of” unless it is clear from the context that this is not intended.

[0041] BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 shows an illustration of an exemplary way of performing the calorimetric method of the present document showing the three different hierarchical levels of analysis.

[0043] Figure 2 shows metabolic profiles of mixed microbial samples for polymicrobial analysis made of i) Staphylococcus aureus + Pseudomonas aeruginosa and ii) Staphylococcus aureus + Candida albicans. See Example 1 .

[0044] Figure 3 shows metabolic profiles of poly similis microbial samples for polymicrobial analysis made of i) Staphylococcus aureus + Corynebacte um striatum and for monomicrobial analysis ii) Staphylococcus aureus. See Example 2.

[0045] Figure 4 shows metabolic profiles of poly similis microbial samples made of i) Pseudomonas aeruginosa + Escherichia coli and for monomicrobial analysis ii) Pseudomonas aeruginosa. See Example 3.

[0046] Figure 5 shows metabolic profiles of poly similis and poly mixtum microbial samples made of i) Candida glabrata + Candida albicans and ii) Candida albicans + Pseudomonas. See Example 4.

[0047] Figure 6 shows the calorimetric analysis using the calorimetric method of the present document of a clinical microbial sample with unknown microbial content. Using the calorimetric method of the present document it was possible to determine that the polymicrobial sample included Enterococcus faecalis and Klebsiella pneumoniae. See Example 5. DETAILED DESCRIPTION

[0048] Overview

[0049] In many situations, such as in case of an infection in a subject, it is of utmost importance to be able to quickly determine if a sample is polymicrobial or monomicrobial, e.g. to direct antimicrobial treatment. Further, it may be important to be able to determine the taxonomic identity of the microorganism(s) in the sample. Traditionally, such analysis has been performed by methods relying on the growth of the microorganisms in the sample on selective media or by using selective dies and observing the staining patterns of the microorganisms in microscopes. However, such prior art methods based on growth of microorganisms or visual observations are time consuming and may not always provide a reliable result. Also, prior art methods do not allow analysis of polymicrobial samples in real time in liquid culture.

[0050] Calorimetry would therefore be an attractive alternative to analyse microbial samples as it allows a quicker analysis of microorganisms. However, thermograms obtained from polymicrobial samples are difficult to analyse as the metabolism of one microorganism may be affected by the other microorganism(s) present in the sample and as a thermogram always represents an aggregation of the individual contribution of the different microorganisms to the calorimetric signal. Thus, even if calorimetry has been demonstrated to be useful to detect the presence of microorganism(s) in a sample and, for monomicrobial samples, in some cases also the taxonomic identity of the microorganism, it has hereinto not been possible to successfully use calorimetry to deduce if a sample with an unknown microbial community is monomicrobial or polymicrobial.

[0051] The present inventors were surprisingly able to develop a calorimetric method for determining if a sample is monomicrobial or polymicrobial.

[0052] The present document is thus directed to a calorimetric method for determining if a sample is a polymicrobial sample comprising a combination of at least two different types of microorganisms (herein also denoted poly mixtum), a polymicrobial sample comprising a combination of at least two different microorganisms of the same type but of different taxonomic identities (herein also denoted poly similis) or a monomicrobial sample. Further, the present method may be used to determine the taxonomic identity of the microorganism(s) in the microbial sample. This method is based on the insight that in order to be able to determine if a microbial sample is monomicrobial or polymicrobial and to determine the type and / or taxonomic identity of the microorganisms in the sample, the calorimetric signal needs to be split up into its different microbial components. However, this turned out to be challenging as the calorimetric signal from a polymicrobial sample is not simply an additive signal of the individual microbial species in the sample, but the microorganisms in a polymicrobial sample may also affect each other so that the total calorimetric signal at any given time point can be higher or lower than expected if the microorganisms would be studied in isolation. This problem was solved by utilizing a carefully selected set of different media during the calorimetric measurements. The method utilizes a set of metabolically differentiating inoculation media that allow the metabolic signal (as determined by calorimetry) from microorganism(s) of different types and / or different taxonomic identities to be split up so that the type(s) of microorganism(s) and optionally the taxonomic identity can be determined by calorimetry.

[0053] The present document is therefore directed to a calorimetric method for the analysis of a microbial sample in order to determine if said microbial sample is:

[0054] I) a polymicrobial sample comprising at least two different types of microorganisms; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities, wherein said types of microorganisms are selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, said method comprising the steps of: a) providing a microbial sample; b) providing a set of metabolically differentiating inoculation media configured to split up a metabolic profile into the different types of microorganisms present in the microbial sample, wherein said set of metabolically differentiating inoculation media comprises a first metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-positive bacterium, a second metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-negative bacterium, and a third metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one fungus; c) inoculating and incubating said microbial sample in said first, second and third metabolically differentiating inoculation medium; d) obtaining calorimetric data from the inoculated and incubated sample of step c), thereby obtaining a metabolic profile from the first, second and / or third metabolically differentiating inoculation medium depending on what types of microorganisms are present in the microbial sample; e) based on the metabolic profiles obtained in step d), determining that said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms if a metabolic profile is obtained in at least two of said first, second and third metabolically differentiating inoculation medium; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if a metabolic profile is obtained in only one of said first, second and third metabolically differentiating inoculation medium. Thus, in accordance with this method, first metabolically differentiating inoculation medium is configured to provide a metabolic profile if said microbial sample comprises at least a Gram-positive bacterium but not if it only comprises Gram-negative bacteria and / or fungi, the second metabolically differentiating inoculation medium is configured to provide a metabolic profile if said microbial sample comprises at least a Gram-negative bacterium but not if it comprises only Gram-positive bacteria and / or fungi, and the third metabolically differentiating inoculation medium is configured to provide a metabolic profile if said microbial sample comprises at least a fungus but not if it comprises only Gram-positive and / or Gramnegative bacteria.

[0055] According to the present calorimetric method, a microbial sample potentially containing one or more types of microorganisms (herein denoted a microbial sample) is inoculated in different metabolically differentiating inoculation media that allow metabolic differentiation between microorganisms of different taxonomic identity. The inoculated sample is then placed in a calorimeter and the calorimetric signal followed to obtain calorimetric data while incubating the microbial sample in the respective metabolically differentiating inoculation media. The calorimetric signal may be followed in real time by so called real-time isothermal (micro)calorimetry. The calorimetric data are used to obtain a metabolic profile of the microbial sample in the respective metabolically differentiating inoculation medium. A calorimetric signal indicates metabolic activity (the signal is therefore a metabolic signal) of one or more microorganisms in the respective metabolically differentiating inoculation media. Depending on in which of the metabolically differentiating inoculation media a calorimetric signal is detected in (and consequently which media a metabolic profile can be obtained from), and in which metabolically differentiating inoculation media no calorimetric signal is detected (and consequently which media no metabolic profile can be obtained from) it can be determined if the sample is i) polymicrobial with at least two different types of microorganisms (i.e , a combination of at least two different types of microorganisms from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi) or ii) monomicrobial or polymicrobial with at least two different microorganisms of the same type but of different taxonomic identities. Thus, if the microbial sample is not polymicrobial with different types of microorganisms, it is determined that the sample is either monomicrobial or polymicrobial with the same type of microorganisms but of different taxonomic identities (i.e., at least two different Gram-positive bacteria, at least two different Gram-negative bacteria or at least two different fungi). By studying the metabolic profiles from the different metabolically differentiating inoculation media, it is also possible to obtain further taxonomic information regarding the microorganism(s) present in the sample, such as family, genus and even species. Thus, depending on the level of detail of the contents of the microbial sample that is desired, different analyses of the calorimetric features may be performed.

[0056] The present method utilizes calorimetric data obtained from a carefully selected set of different metabolically differentiating inoculation media. The metabolically differentiating inoculation media are selected based on their abilities to promote or suppress the metabolic activity of microorganisms of different type and / or different taxonomic identity (e g. fungi vs bacteria, Gram-positive bacteria vs Gram-negative bacteria, different families, genera, species). The media used in the present document are therefore denoted “metabolically differentiating inoculation media”. Such media may herein also be denoted a “medium” or “media”. Some of the metabolically differentiating inoculation media contain components that allow for selection and / or differentiation of metabolic activity between different types of microorganisms (i.e. between Gram-positive bacteria, Gram-negative bacteria and fungi), while others of the metabolically differentiating inoculation media are more general and support metabolic activity of more than one type of microorganism. If a calorimetric signal is detected after incubation in a specific type of metabolically differentiating inoculation medium, this indicates the presence of metabolic activity and thus the presence of microorganism(s) of certain type / taxonomic identity.

[0057] The metabolic activity can be detected by calorimetry as a heat signal. Microbial cells can be metabolically active without concomitant growth and an increase in metabolic activity often precedes growth. Due to the use of calorimetry and the specially selected metabolically differentiating inoculation media, a mandatory dependency on growth of microorganisms for the testing is therefore avoided, which leads to that a result may be obtained faster since only metabolic activity of the microbial cells in the microbial sample is necessary to detect a calorimetric signal Further, as only viable cells are able to be metabolically active, the calorimetric method is not disturbed by the potential presence of dead microorganisms. These are advantageous features over prior art methods for studying microbial communities in liquid culture as such methods are mandatorily dependent on growth and as dead and / or inactive cells may skew the results.

[0058] Incubation of the microbial samples in the calorimeter

[0059] The microbial sample is inoculated in the different metabolically differentiating inoculation media and the microbial sample is incubated in a calorimeter and the metabolic activity determined as heat flow registered by the calorimeter to obtain a metabolic profile of the microbial sample in each metabolically differentiating inoculation medium.

[0060] The temperature during the incubation steps of the present calorimetric method will differ depending on the type of sample. For example, an environmental sample may be incubated at a lower temperature than a clinical sample, depending on what microorganisms are expected to be present in the sample. A clinical sample is typically incubated at a temperature of from about 35 °C, to about 39 °C, such as from about 36 °C, to about 38 °C, such as about 37 °C. An environmental sample may be incubated at a temperature from about 10 °C to about 30 °C, such as from about 15 °C to about 25 °C. However, these temperature ranges are only exemplary and other temperatures may be used depending on the microbial sample to be analysed.

[0061] Typically, the incubation is performed for a time period of from about 1 hour to about 48 hours as counted from the time to detection (i.e., the time point when a calorimetric signal first appears in the respective metabolically differentiating inoculation medium).

[0062] The microbial sample may be incubated under both aerobic and anaerobic conditions in order to allow metabolic activity from both aerobic and anaerobic microorganisms to be detected. Alternatively, the microbial sample may be incubated under only aerobic or anaerobic conditions This may for example be suitable if it is expected that no anaerobic or aerobic microorganisms, respectively, are present in the microbial sample.

[0063] Optional pre-incubation and re-loading

[0064] The calorimetric method of the present document may comprise an optional step of preincubating and reloading the samples before obtaining the metabolic profiles used for the analysis of the microbial sample.

[0065] The calorimetric method of the present document may thus comprise a pre-incubation and re-loading step before inoculating and incubating the microbial sample in the metabolically differentiating inoculation media. This pre-incubation and re-loading step comprises inoculating and incubating the microbial sample in at least one inoculation medium.

[0066] The inoculation medium for this pre-incubation may be one or more of the metabolically differentiating inoculation media disclosed elsewhere herein or a different inoculation medium than the metabolically differentiating inoculation media, such as a general purpose growth medium known to the person skilled in the art. Preferably, the inoculation medium used for the pre-incubation step is a general purpose inoculation medium supporting growth of a large variety of microbial types and taxonomic identities so that the microbial population after pre-incubation comprises the same mixture of microorganisms as the original microbial sample.

[0067] The microbial sample(s) is thus inoculated in at least one inoculation medium and the metabolic activity followed calorimetrically. When the R value for the metabolic rate (i.e. metabolic activity over time) is >0.98 for at least about 20 min or when the metabolic activity of the pre-incubated sample reaches the inflection point or within about 2 hours, such as within about 1 hour, after the inflection point is reached, one or more aliquot(s) of the pre-incubated sample(s) is reloaded in the first, second and third metabolically differentiating inoculation medium.

[0068] According to the present calorimetric method, the metabolic activity is thus followed over time to identify when the metabolic rate reaches an R value >0.98. The reloading should be performed when the metabolic rate has had an R value >0.98 over a time period of preferably at least about 20 minutes, preferably for at least 30 minutes, at the inflection point or within two hours after the inflection point has been reached. This ensures that the microbial population is sufficiently homogenous as regards metabolic activity and that the sample is in a qualitatively suitable state that facilitates the determination of the type and / or taxonomic identity of the microorganisms therein. Alternatively, instead of applying a reloading step, different amounts of the microbial sample may be inoculated and incubated in step c) to ensure that at least some of the inoculated and incubated samples contain enough microbes for a calorimetric signal to be detected.

[0069] The inflection point for metabolic activity as defined herein signifies the inflection from a linear increase in metabolic rate R >0.98 to when it is dropping below R <0.98. For determination of the inflection point, the linear increase in metabolic activity, i.e. when R >0.98, should preferably continue for some time before dropping below R <0.98. Typically, the linear increase in metabolic activity should continue for at least 10-30 min before dropping to R <0.98 when determining the inflection point. Preferably, the linear increase continues for at least 20 min or at least 30 min. The inflection point may be determined from a thermogram by identifying the point where the increase in metabolic activity over time is the highest. Depending on the microbial species in the microbial sample, it may still be possible to reload at a time point more than two hours after the inflection point is reached.

[0070] Pre-incubating the microbial sample and reloading it around the inflection point as disclosed herein may facilitate the analysis of the calorimetric data after incubation in the different metabolically differentiating inoculation media.

[0071] Thus, the calorimetric method of the present document may comprise a pre-incubation and reloading step. Such a calorimetric method for the analysis of a microbial sample in order to determine if said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities, wherein said types of microorganisms are selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, may comprise the steps of: a) providing a microbial sample; b) providing a set of metabolically differentiating inoculation media configured to split up a metabolic profile into the different types of microorganisms present in the microbial sample, wherein said set of metabolically differentiating inoculation media comprises a first metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-positive bacterium, a second metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gramnegative bacterium, and a third metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one fungus; c) inoculating and incubating the microbial sample in at least one inoculation medium and calorimetrically following the metabolic activity of the incubated sample and when the R value for the metabolic rate is >0.98 for at least about 20 min or when the metabolic activity of the incubated sample reaches the inflection point or within 2 hours after said inflection point is reached, reloading one or more aliquot(s) of said incubated sample in said first, second and third metabolically differentiating inoculation medium and incubating the sample in said first, second and third metabolically differentiating inoculation media; d) obtaining calorimetric data from the sample of step c) inoculated and incubated in said first, second and third metabolically differentiating inoculation medium, thereby obtaining a metabolic profile from the first, second and / or third metabolically differentiating inoculation medium depending on what types of microorganisms are present in the microbial sample; e) based on the metabolic profiles obtained in step d), determining that said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms if a metabolic profile is obtained in at least two of said first, second and third metabolically differentiating inoculation medium; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if a metabolic profile is obtained in in only one of said first, second and third metabolically differentiating inoculation medium.

[0072] Such a re-loading step may also be used if a metabolic profile is obtained from further media for the analysis at the second and / or third hierarchical analysis level(s).

[0073] It is also possible to reload two or more aliquots with different total metabolic activity after the pre-incubation (e.g. by re-loading different volumes of the pre-incubated samples) and then use the calorimetric data from the aliquot(s) corresponding to the aliquot(s) wherein the metabolic activity in the sample(s) directly after reloading is below the detection limit of a calorimetric signal and the metabolic activity in the same aliquot(s) is detected between about 15 min and about 6 hours, such as between about 15 min and about 3 hours, after said reloading. Typically, a couple of different aliquots are transferred in the re-loading step so that from about minimum 0.1 % to maximum 50% (such as from about 1 % to about 15%, such as from 1 % to about 10% or from about 5% to about 10%) of the metabolic activity of the incubated sample is reached in the reloaded sample at the time of reloading. Le., if the metabolic activity in the incubated sample at the time of reloading is X J / ml, then a metabolic activity corresponding to from about 0.1% to 50% of X J / ml is present in the reloaded sample directly after reloading. Preferably, at least two aliquots are transferred to reach different metabolic starting activities in the reloaded sample in the different metabolically differentiating inoculation media.

[0074] A computer and software may be used to determine the time point for reloading and / or which aliquots to use for analysis.

[0075] Metabolically differentiating inoculation media and analysis of metabolic profiles

[0076] In order to be able to split up the metabolic activity from a microbial sample so that it is possible to calorimetrically determine the type and / or taxonomic identity of the microorganism(s) potentially present in the microbial sample, the microbial sample is divided up and inoculated in a set of different metabolically differentiating inoculation media. The set of metabolically differentiating inoculation media comprises at least one first metabolically differentiating inoculation medium, at least one second metabolically differentiating inoculation medium and at least one third metabolically differentiating inoculation medium. The set may comprise more than one of the first, second and third metabolically differentiating inoculation medium, respectively. These sets of metabolically differentiating inoculation media have been carefully selected to device a scheme of how a polymicrobial metabolic activity can be detected so the identity of the microorganism(s) potentially present in the microbial sample can be determined. If a metabolic activity is not detected in any of the metabolically differentiating inoculation media used (i.e. no metabolic profile obtained), it can be determined that the microbial sample most likely does not comprise any microorganisms as the metabolically differentiating inoculation media are selected so that virtually all microorganisms should exhibit metabolic activity in at least one of the metabolically differentiating inoculation media. The metabolically differentiating inoculation media used in accordance with the present document are optionally supplemented with compounds that either select for and / or differentiate between microorganism of different types and / or taxonomic identities, in order to increase the specificity of the method. Such compounds are described further elsewhere herein. Exemplary metabolically differentiating inoculation media for splitting up the metabolic signal in order to be able to determine the type(s) and / or taxonomic identity / identities of the microorganism(s) in a microbial sample are presented in Table 1 below. The metabolically differentiating inoculation media may further be supplemented with compounds (i.e. supplements) increasing the selective and / or differentiating ability of the respective metabolically differentiating inoculation medium. The metabolically differentiating inoculation media and supplements disclosed herein may be replaced with other media and / or supplements allowing the same purpose to be achieved. The present calorimetric method is therefore not limited to the use of the specific media and supplements exemplified herein. The calorimetric method of the present document is further illustrated in a non-limiting way in Fig. 1 , where exemplary ways of performing the analysis at the different hierarchical levels (1st, 2ndand 3rd) can be seen.

[0077] Table 1 Exemplary media to determine the type and / or identity of different types of microorganisms.

[0078] Analysis at a first hierarchical level

[0079] At a first hierarchical level, a first set of metabolically differentiating inoculation media is used to determine if the microbial sample is:

[0080] -i) a polymicrobial sample comprising at least two different types of microorganisms; or -ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type.

[0081] At this first hierarchical level it can thus be directly determined if the microbial sample is a polymicrobial sample with at least two different types of microorganisms (poly mixtum), i.e. that the microbial sample comprises at least to different types of microorganisms selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi. The set of metabolically differentiating inoculation media used for the determination at the first hierarchical level comprise at least three different metabolically differentiating inoculation media (i.e. at least one for each type of microorganism) configured to split up the metabolic profiles into the different types of microorganisms potentially present in the microbial sample. More than one metabolically differentiating inoculation medium specific for each type of microorganism may be used.

[0082] The first metabolically differentiating inoculation medium is configured to provide a metabolic profile if the microbial sample comprises at least one Gram-positive bacterium but not if it only comprises Gram-negative bacteria and / or fungi. For example, the first metabolically differentiating inoculation medium may selected from the group consisting of salt mannitol broth and CLED broth. The first metabolically differentiating inoculation medium may be supplemented with one or more selective and / or differentiating supplements that promotes metabolic activity from Gram-positive bacteria and / or that inhibits metabolic activity from Gram-negative bacteria and / or fungi. For example, the supplement for the first metabolically differentiating inoculation medium may comprise one or more selective and / or differentiating supplements selected from the group consisting of salts, such as sodium chloride and / or lithium chloride; antibiotics, such as colistin polymyxin B; and / or antimicrobials, such as phenyl ethanol. However, other media and supplements enabling a metabolic profile to be obtained if the microbial sample comprises at least one Gram-positive bacterium but not if it only comprises Gram-negative bacteria and / or fungi may be used instead or in addition to the first metabolically differentiating inoculation media and supplements exemplified herein.

[0083] The second metabolically differentiating inoculation medium is configured to provide a metabolic profile if the microbial sample comprises at least one Gram-negative bacterium but not if it comprises only Gram-positive bacteria and / or fungi. The second metabolically differentiating inoculation medium may be selected from the group consisting of MacConkey broth, Hektoen enteric agar, and Sabouraud broth. The second metabolically differentiating inoculation medium may be supplemented with one or more selective and / or differentiating supplements that promotes metabolic activity from Gramnegative bacteria and / or that inhibits metabolic activity from Gram-positive bacteria and / or fungi. The supplement for the second metabolically differentiating inoculation medium may comprise one or more selective and / or differentiating supplements, such as supplements selected from the group consisting of salts, such as eosin Y; sodium taurocholate; sodium selenite; antibiotics, such as vancomycin, cycloheximide, and / or azinomycin; antimicrobials such as brilliant green; and / or saccharides such as lactose and / or glucose. However, other media and supplements enabling a metabolic profile to be obtained if the microbial sample comprises at least one Gram-negative bacterium but not if it only comprises Gram-positive bacteria and / or fungi may be used instead or in addition to the second metabolically differentiating inoculation media and supplements exemplified herein.

[0084] The third metabolically differentiating inoculation medium is configured to provide a metabolic profile if the microbial sample comprises at least one fungus but not if it comprises only Gram-positive and / or Gram-negative bacteria The third metabolically differentiating inoculation medium may be selected from the group consisting of Sabouraud broth, malt extract broth, and yeast extract-peptone dextrose broth. The third metabolically differentiating inoculation medium may be supplemented with one or more selective and / or differentiating supplements that promotes metabolic activity from fungi and / or that inhibits metabolic activity from Gram-positive bacteria and / or Gram-negative bacteria. The supplement for the third metabolically differentiating inoculation medium may comprise one or more selective and / or differentiating supplements selected from the group consisting of antibiotics such as gentamycin, chloramphenicol, cephalothin, and cefamandole. However, other media and supplements enabling a metabolic profile to be obtained if the microbial sample comprises at least one fungus but not if it only comprises Gram-positive bacteria and / or Gram-negative bacteria may be used instead or in addition to the third metabolically differentiating inoculation media and supplements exemplified herein.

[0085] Depending on in which of the first, second and third metabolically differentiating inoculation medium is detected it can be determined which type(s) of microorganisms are present in the sample. If metabolic activity is detected in the metabolically differentiating inoculation medium specific for Gram-positive bacteria and the fungi specific medium but not in the Gram-negative specific medium, it can be determined that the microbial sample comprises a combination of Gram-positive bacteria and fungi. If metabolic activity is detected in the Gram-positive specific medium and the Gramnegative specific medium but not in the fungus-specific medium, it can be determined that the microbial sample comprises a combination of Gram-positive bacteria and Gramnegative bacteria. If metabolic activity is detected in the Gram-negative specific medium and fungus-specific medium but not in the Gram-positive specific medium, it can be determined that the microbial sample comprises a combination of Gram-negative bacteria and a fungus If metabolic activity is detected in the Gram-positive specific medium, the Gram-negative specific medium and the fungus-specific medium, it can be determined that the microbial sample comprises a combination of Gram-positive bacteria, Gram-negative bacteria and fungi. If metabolic activity is detected in only one of the metabolically differentiating inoculation media, it can be determined that the microbial sample does not contain different types of microorganisms (even if the microbial sample may contain two or more microorganisms of different taxonomic identity but being of the same type). As earlier mentioned, if no metabolic activity is detected in any of the first, second and third metabolically differentiating inoculation medium, it can be determined that the microbial sample most likely does not contain any microorganisms.

[0086] Thus, if a metabolic profile is obtained in two or more of the metabolically differentiating inoculation media at the first hierarchical level, it can be determined that the microbial sample comprises at least two different types of microorganisms and also which these types of microorganisms are. The microbial sample may of course comprise more than one microorganism of each type, i.e. two or more microorganisms of the same type but of different taxonomic identities. However, at the first hierarchical level, it is only determined if the microbial sample comprises at least two different types of microorganisms and not if two or more taxonomically different microorganisms of the same type constitute the type(s) detected. This can be determined at the third hierarchical level as explained below.

[0087] Being able to quickly determine if a microbial sample contains a mixture of different types of microorganisms can be of uttermost importance in a clinical situation with a patient having an infection of unknown origin. By knowing that a sample is polymicrobial, the right treatment can be set on much faster, which improves the treatment outcome for the patient. The present calorimetric method allows a much quicker result regarding if a microbial sample is polymicrobial being obtained than has been possible with previously employed methods.

[0088] Should only one or two of the three metabolically differentiating inoculation media in the first set of metabolically differentiating inoculation media be used, it would still be possible to determine that a sample is polymicrobial if a metabolic profile is obtained from both metabolically differentiating inoculation media used However, should the microbial sample contain a microorganism of the type for which a metabolically differentiating inoculation media is not used, then the microbial sample may be mistaken to be a monomicrobial sample or it would not be possible to determine that the microbial sample contains all three types of microorganisms. Thus, the first set of metabolically differentiating inoculation media preferably comprises a first, second and third metabolically differentiating inoculation medium according to the above.

[0089] Analysis at a second hierarchical level

[0090] As mentioned above, if a microbial sample comprises at least two different types of microorganisms, this may be determined already at the first hierarchical level using the first set of metabolically differentiating inoculation media. If it at the first hierarchical level is determined that the microbial sample does not comprise at least two different types of microorganisms, it can be concluded that the microbial sample comprises at least two microorganisms having different taxonomic identities but being of the same type (poly similis) or that the microbial sample is a monomicrobial sample.

[0091] The microbial sample may in this case be analysed at a second hierarchical level to distinguish between these two cases (i.e at least two microorganisms of the same type but having different taxonomic identities or a monomicrobial sample). This second hierarchical level analysis comprises inoculating and incubating the microbial sample in different metabolically differentiating inoculation media and analysing the metabolic profile(s) obtained with metabolic profiles previously obtained for known monomicrobial and / or polymicrobial samples incubated in the same metabolically differentiating inoculation media as used at the second hierarchical level.

[0092] The present calorimetric method may thus comprise a further a step of determining that said microbial sample is:

[0093] I) a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if said metabolic profile(s) match(es) two or more monomicrobial metabolic profiles or matches a combined metabolic profile of a polymicrobial sample with known microbial content; or ii) a monomicrobial sample if the metabolic profile(s) match(es) a monomicrobial metabolic profile, wherein the metabolic profile(s) of the microbial sample are obtained from: a. at least one metabolically differentiating inoculation medium of the same type as the first, second or third metabolically differentiating inoculation medium in which a metabolic profile was obtained according to step e) ii) for the analysis at the first hierarchical level; and b. at least one non-selective general-purpose inoculation medium, such as tryptic soy broth or Mueller Hinton broth.

[0094] The calorimetric analysis according to the second hierarchical level comprises inoculating and incubating the microbial sample in at least one metabolically differentiating inoculation medium of the same type as the metabolically differentiating inoculation medium / media from which a metabolic profile was obtained at the first hierarchical level (i.e. a first, second or third metabolically differentiating medium as defined above). Further, the microbial sample may be inoculated and incubated in a fourth type of metabolically differentiating inoculation medium, which is a non-selective general-purpose inoculation medium, such as tryptic soy broth or Mueller Hinton broth, which enables a metabolic profile to be obtained from most microorganisms. This fourth metabolically differentiating inoculation medium is therefore not selective for any specific type of microorganism. The set of metabolically differentiating inoculation media comprising a first, second, and / or third and a fourth metabolically differentiating inoculation medium for use for analysis at the second hierarchical level may herein be denoted a second set of metabolically differentiating inoculation media.

[0095] The first metabolically differentiating inoculation medium and supplements for analysis at the second hierarchical level are therefore of the same type as defined for the first metabolically differentiating inoculation medium in the first hierarchical and non-limiting examples of such metabolically differentiating media are given above. The second metabolically differentiating inoculation medium and supplements for analysis at the second hierarchical level are therefore of the same type as defined for the second metabolically differentiating inoculation medium in the first hierarchical and non-limiting examples of such metabolically differentiating media are given above. The third metabolically differentiating inoculation medium and supplements for analysis at the second hierarchical level are therefore of the same type as defined for the third metabolically differentiating inoculation medium in the first hierarchical and non-limiting examples of such metabolically differentiating media are given above. A further example of such a third metabolically differentiating inoculation medium for analysis at the second hierarchical level is Czapek Dox broth. The fourth metabolically differentiating inoculation medium in the second set of metabolically differentiating inoculation media is a general- purpose medium. Examples of such media include, but are not limited to, tryptic soy broth and Mueller Hinton broth. The fourth metabolically differentiating inoculation medium typically does not contain any selective and / or differentiating supplements. For a determination at the second hierarchical level, the microbial sample is inoculated and incubated in the chosen metabolically differentiating inoculation media and a metabolic profile obtained as described for the first hierarchical level. The metabolically differentiating inoculation media may be the same as the ones used in the first hierarchical level or different. It is thus also possible to use the metabolic profile(s) obtained at the first hierarchical level for the analysis at the second hierarchical level and / or further metabolic profiles may be obtained by inoculating and incubating in one or more further metabolically differentiating inoculation media.

[0096] For an analysis at the second hierarchical level to determine if a microbial sample is a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities or a monomicrobial sample, the obtained metabolic profiles are compared to metabolic profiles of microbial samples with known microbial content, preferably obtained from the same metabolically differentiating inoculation media as the microbial sample were inoculated in. These metabolic profiles may be obtained from polymicrobial and / or monomicrobial samples. Thereby, it can be determined that a microbial sample is a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if the metabolic profile(s) match(es) two or more monomicrobial metabolic profiles or matches a combined metabolic profile of a polymicrobial sample with known microbial content. Alternatively, it can be determined that the microbial sample is a monomicrobial sample if the metabolic profile(s) match(es) a monomicrobial metabolic profile.

[0097] Analysis at a third hierarchical level

[0098] The microbial sample may further be analysed at a third hierarchical level, wherein the family, genus and / or species of the microorganisms present in the microbial sample can be determined. Typically, the third hierarchical level analysis is performed to determine the taxonomic identity of the microorganism(s) present in the microbial sample, i e. their family, genus and / or species.

[0099] For analysis at the third hierarchical level, metabolic profiles are obtained by inoculating and incubating a microbial sample in one or more metabolically differentiating inoculation media as described herein for the first and second hierarchical levels. The metabolic profiles obtained are then analysed by comparing them to the metabolic profiles obtained from monomicrobial and / or polymicrobial samples with known microbial content, as explained for the analysis at the second hierarchical level. The set of metabolically differentiating inoculation media for use for analysis at the third hierarchical level may herein be denoted a third set of metabolically differentiating inoculation media.

[0100] Exemplary media for analysis at the third hierarchical level are given in Table 1 above. Further, the first, second and / or third metabolically differentiating inoculation media and their different supplements as described under the first and second hierarchical level above and / or a general-purpose medium as described for the second hierarchical level may be used to obtain the metabolic profile(s) for analysis at the third hierarchical level. It is thus also possible to use the metabolic profile(s) obtained in the first and / or second hierarchical level(s) for the analysis at the third hierarchical level and / or further metabolic profiles may be obtained by inoculating and incubating in one or more further metabolically differentiating inoculation media.

[0101] For all hierarchical levels of analysis, metabolic profiles used for comparison with a metabolic profile obtained from the microbial sample to be analysed are typically obtained under the same conditions (same metabolically differentiating inoculation medium, same supplements, same temperature etc.) as the metabolic profiles obtained from the unknown microbial sample to facilitate the analysis. Depending on which microorganism(s) are present in the microbial sample, one or more metabolic profiles obtained under different conditions (such as in different metabolically differentiating inoculation media) may have to be analysed.

[0102] Although the present method is described in sequential steps of analysis for the first, second and third hierarchical level, in practise, often the analysis at the different hierarchical levels is performed simultaneously, in particular when a computer-assisted analysis is performed (see below). The sequential description of the method steps of the method may therefore be more for illustrative purposes than for practical purposes. According to the present document it is possible to perform an analysis at the first, second and / or third hierarchical level, i.e it is not necessary to always perform an analysis at all hierarchical levels. At which hierarchical level a microbial sample is analysed will depend upon which information about the content of a microbial sample that is desired. An analysis at the second or third hierarchical level will of course also provide information regarding if the sample is a monomicrobial sample or a polymicrobial sample.

[0103] Calorimetric data, calorimetric feature(s) and metabolic profiles

[0104] For the analysis at the second and / or third hierarchical level, previously obtained metabolic profiles created by inoculating and incubating microorganisms of known taxonomic identity in different metabolically differentiating inoculation media are used. Typically, the microorganisms used to create these metabolic profiles have been inoculated and incubated in a similar manner as the microbial sample to be tested, e.g. at the same temperature and in the same metabolically differentiating inoculation medium as described elsewhere herein. Typically, the previously established metabolic profiles are created by inoculating and incubating both monomicrobial cultures and polymicrobial cultures in different metabolically differentiating inoculation media as the metabolic profile of a combination of microorganisms may differ from the aggregated metabolic profile obtained from a monomicrobial culture as explained elsewhere herein. However, for the calorimetric method of the present document to work it is not necessary to have a metabolic profile of every combination of microorganisms that possibly could occur as the metabolically differentiating inoculation media are configured to split up the metabolic signal from a polymicrobial sample into its monomicrobial constituents. Including metabolic profiles of combinations of microorganisms of different taxonomic identities may however facilitate the analysis of metabolic profiles and enable a quicker and / or more accurate result to be obtained.

[0105] As explained above, by calorimetry, the heat flow in a sample can be measured and this heat flow reflects the metabolic activity of microbial cells in the sample. Heat flow is defined as the measured signal in isothermal calorimetry and is expressed as energy / time unit, e.g. J / s.

[0106] In the calorimetric method of the present document, the metabolic activity of the incubated samples is thus detected by calorimetry. A higher metabolic activity results in a higher calorimetric signal. The calorimetric signal in a sample over time is usually displayed in a so called thermogram. Such a thermogram may in itself represent a metabolic profile.

[0107] However, the calorimetric data in a calorimetric thermogram can also be transformed into a set of calorimetric features that can be used for the analysis and determination of the type and taxonomic identity of the microorganism(s) in the sample, see examples of such features in Table 2.

[0108] In Table 2, an overview of different exemplary calorimetric features, obtained by analysis of calorimetric data, that may be used in obtaining a metabolic profile are presented It is possible to use any one or any combination of two or more of these calorimetric features to obtain a metabolic profile according to the present document. Depending on where differing calorimetric feature(s) can be found between different microorganisms, different calorimetric features or different combinations of calorimetric features can be used in order to enhance the differences in the metabolic activity from different microorganisms so that the metabolic profile obtained from a specific metabolically differentiating inoculation medium can be split up into its monomicrobial components. Thus, by selecting the metabolically differentiating inoculation media and the calorimetric features analysed so that they enhance the differences in metabolic profiles between different microorganisms, it is possible to split up the metabolic signal in a polymicrobial sample to determine not only that the sample is polymicrobial but also what combination of microorganisms are present in it. 1. Time to peak.

[0109] 2. Amplitude of highest peak with a length of > 1h and a prominence of > 1% of the highest peak amplitude.

[0110] 3. Total area under the curve (sum of heat generated)

[0111] • Same as above but for 1st, 2nd, 3rd, 4thtime interval.

[0112] • Same as above but relative 1st, 2nd, 3rd, 4thtime interval

[0113] 4. Number of peaks with a prominence of at least 2% of the highest peak amplitude

[0114] • Same as above but for 1st, 2nd, 3rd, 4th, 5thtime interval.

[0115] • Same as above but for prominence 7,5%, 15%, 40%, 50% of highest peak amplitude and width 1h.

[0116] • Same as above but with width 2h, 4h, 8h.

[0117] 5. Exponential vs decay ratio

[0118] • From highest peak, descend left until 90% of peak value call this point (L_P90), keep descending until at 25% of peak value call this point (L_P25).

[0119] • From highest peak, descend right until 90% of peak value call this point (R_P90), keep descending until at 25% of peak value call this point (R_P25).

[0120] • Calculate L_AUC as area under curve from (L_P25) to (L_P90).

[0121] • Calculate R_AUC as area under curve from (R_P25) to (R_P90).

[0122] • Calculate exponential vs decay ratio as (L_AUC) / (L_AUC+ R_AUC).

[0123] 6. Percentiles 5, 25, 50, 75, 95.

[0124] 7. Percentiles (85) - Percentiles (15).

[0125] 8. Arrays

[0126] • Multiple peaks with prominence 1 - 10 % of maximum peak value and width 2-10 h.

[0127] 9. Twin peaks detector

[0128] • Looks for two highest peaks above 5% prominence of the highest peak amplitude, that are not overlapping at least for 2h.

[0129] Table 2. Exemplary calorimetric features that may be used to obtain a metabolic profile.

[0130] When a computer is used for analyses, often all of the above exemplified calorimetric features are analysed It is also possible to use additional calorimetric features known to the person skilled in the art for the analysis.

[0131] For an analysis, particularly at the second and / or third hierarchical level, metabolic profiles previously obtained from known microorganisms incubated under the same conditions are used by either direct comparison with the metabolic profiles of the microbial sample, or indirectly by encoding the metabolic profiles of the microbial sample and the known microorganisms into a model. Thus, the metabolic profiles obtained from known monomicrobial and polymicrobial samples may be stored in a database as a library of metabolic profiles and used for comparison with metabolic profiles obtained from microbial samples with unknown content or it is also possible to train a model using the previously obtained profiles and use that model to classify unknown microbial samples. Thus, in order to establish the type and / or taxonomic identity of microorganism(s) in a microbial sample, the metabolic profiles are matched with previously obtained metabolic profiles obtained from polymicrobial and / or monomicrobial samples with known microbial content.

[0132] The previously obtained metabolic profiles are preferably obtained from both monomicrobial samples and polymicrobial samples. This is important as, as explained elsewhere herein, the metabolic profile of a combination of microorganisms in a polymicrobial sample often will not simply be the combined profile of the individual metabolic profiles of the monomicrobial components, but will differ as microorganisms incubated together will affect each other’s metabolic profiles. Using metabolic profiles obtained from different combinations of microorganisms (i.e., from polymicrobial samples) is therefore important in order to be able to better split up a metabolic profile of a polymicrobial sample into its monomicrobial components to determine their type and / or identity.

[0133] The metabolic profiles obtained may thus, particularly at the second and third hierarchical levels of analysis, be analysed by comparing them with a library of previously obtained metabolic profiles of different microorganisms incubated alone or in different combinations in the different metabolically differentiating inoculation media (i.e first, second, third and / or fourth metabolically differentiating inoculation media as defined elsewhere herein) under at least substantially the same conditions as the microbial sample to be analysed. Such an analysis may be performed by visual inspection or by using a computer and a database with the previously obtained metabolic profiles. Such a database may thus be used to establish a correlation between calorimetric feature(s) (making up the metabolic profiles) and a plurality of microorganisms allowing them to be distinguished from each other. A computer and a software may be used for this analysis. The identification may be performed by an identification algorithm matching the metabolic profiles obtained from each differentiating inoculation medium against a database (i.e. a library of metabolic profiles) containing the previously obtained metabolic profiles. The metabolic profiles may thus be analysed using an identification algorithm matching a metabolic profile of a microbial sample against a database containing metabolic profiles of known microorganisms.

[0134] Alternatively, the analysis, particularly at the second and third hierarchical levels may be performed using a supervised and / or unsupervised machine learning models such as Random Forest and / or Rocket trained on monomicrobial and polymicrobial samples obtained as explained above. Calorimetric input data are computed into calorimetric features. For training the model to distinguish between microorganisms of different types and / or taxonomic identities, metabolically differentiating inoculation media which preferentially selects for calorimetric features of one type / taxonomic identity of microorganism are used. The incoming calorimetric features may for example be analyzed using a Random Forest supervised machine learning algorithm trained for species identification. The forest tree model is comprised of a collection of decision trees, built independently and randomly selected from the training data and a subset of calorimetric features. For identification of taxonomic identity, the outputs from the individual trees are combined in order to make a prediction, resulting in an increased robustness compared to the predictions from each individual decision tree. Typically, the 9 main calorimetric feature groups according to the above are used to build the tree to make a prediction of the species identification.

[0135] The present document is therefore also directed to a system for performing a calorimetric method of the present document. The system comprises a calorimeter and a software for performing a calorimetric analysis in the calorimetric method of the present document, such as an analysis of the type and / or taxonomic identity, determination the inflection point and / or for determining which re-loaded sample to analyse.

[0136] The present document is also directed to a computer program comprising computer program code, the computer program code being adapted, if executed on a processor, to implement a calorimetric method of the present document. The present document is therefore also directed to a computer program comprising computer program code, the computer program code being adapted to, if executed on a processor, implement the step(s) of determining of the calorimetric method of the present method to receive calorimetric data, the data being indicative of the inoculated and incubated sample of step c) of the calorimetric method of the present document, thereby obtaining a metabolic profile from a first, second and / or third metabolically differentiating inoculation medium depending on what types of microorganisms are present in the microbial sample; determine, using the received calorimetric data and the obtained metabolic profiles that said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms if a metabolic profile is obtained in at least two of said first, second and third metabolically differentiating inoculation medium; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if a metabolic profile is obtained in only one of said first, second and third metabolically differentiating inoculation medium. The present document is also directed to a computer program product comprising a computer readable storage medium, the computer readable storage medium having this computer program.

[0137] The present document is also directed to a computer-implemented method for determining if a microbial sample is a polymicrobial sample comprising a combination of at least two different types of microorganisms selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi or at least two different microorganisms of the same type selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi in a sample, and optionally determining the taxonomic identity of one or more microorganisms in said microbial sample, said method comprising the steps of: a) receiving calorimetric data from a calorimeter; b) using the calorimetric data to obtain calorimetric features; c) establishing a correlation between said calorimetric features and a plurality of microorganisms; d) determining the type and / or taxonomic identity based on said calorimetric features. As mentioned above, the calorimetric features make up the metabolic profile of a microbial sample.

[0138] The present document also discloses computer implemented method for training a machine learning model to identify one or more microorganisms, the method comprising:

[0139] -obtaining calorimetric feature data, the calorimetric feature data being derived from calorimetric data being indicative of the metabolic activity of the one or more microorganism(s) in a set of different metabolically differentiating inoculation media as defined herein;

[0140] -obtaining media data indicative of the set of different metabolically differentiating inoculation media;

[0141] -obtaining microorganism data indicative of a set of microorganisms from which the one or more microorganisms are identified;

[0142] -generating a trained machine learning model by training, using training data comprising the calorimetric feature data, the media data and the microorganism data, a machine learning model to output, based on input operating data, identification of one or more microorganisms.

[0143] The set of metabolically differentiating inoculation media used for training such a machine learning model may be any set of metabolically differentiating inoculation media disclosed herein, i.e. a set of metabolically differentiating inoculation media used for analysis at the first, second and / or third hierarchical level.

[0144] The present document also discloses a kit for performing a calorimetric method of the present document, said kit comprising:

[0145] I) a set of metabolically differentiating inoculation media as defined elsewhere herein; ii) one or more database(s) which contain metabolic profiles of known microorganisms incubated in the same metabolically differentiating inoculation media (and under at least substantially the same conditions); iii) a software component comprising a function that determines the type and / or taxonomic identity of a microorganism.

[0146] The present document also discloses a kit for performing a calorimetric method of the present document, said kit comprising: i) a set of metabolically differentiating inoculation media as defined elsewhere herein; ii) a training data set which contains metabolic profiles of known microorganisms incubated in the same metabolically differentiating inoculation media; iii) a software component comprising a function that determines the type and / or taxonomic identity of a microorganism.

[0147] The set of metabolically differentiating inoculation media of a kit of the present document comprises at least one first metabolically differentiating inoculation medium, at least one second metabolically differentiating inoculation medium and at least one third metabolically differentiating inoculation medium.

[0148] As mentioned above, the first, second and third hierarchical level of analysis do not have to be performed in their numerical order, but the microbial sample may be analysed at the different levels in any other sequential order or in parallel (simultaneously). In particular, when a computer is used for the analysis of the metabolic profiles, several or all steps in the analysis may be performed simultaneously (i.e.in parallel) by the computer and the information regarding if the sample is mono- or polymicrobial and information regarding the taxonomic identity may be obtained at the same time. For example, obtaining information regarding the family, genus and / or species of the microorganism(s) present in the microbial sample (i.e. an analysis at the third hierarchical level), simultaneously may allow deducing if the sample is mono- or polymicrobial. Depending on the microorganism(s) present in the sample and the level of detail regarding these, the analysis does not have to be performed at all levels.

[0149] The microbial sample

[0150] The microbial sample in accordance with the present document is a sample potentially comprising microorganism(s). The microbial sample may herein also simply be denoted a “sample”. The microbial sample may, e.g , be a clinical sample from a subject, an environmental sample, a food or feed sample, and / or a purified microbiological sample. Typically, the sample is a clinical sample from a human or animal subject. Such a clinical sample may be a sample of any bodily fluid or tissue, such as blood, urine, plasma, synovial fluid, cerebrospinal fluid, bone tissue, soft tissue, connective tissue, skin tissue, sputum, lavage fluids, plasma / serum, urine, peritonea fluid, pericardial fluid, and / or pleural fluid The microbial sample may be a tissue biopsy from a subject.

[0151] A great advantage with the present calorimetric method is that the sample does not have to be purified and / or that the microorganisms do not have to be isolated from the sample. Rather, the sample can be directly inoculated into the metabolically differentiating inoculation media and analysed by calorimetry. This results in that a result is more rapidly obtained than if a method that requires prior purification / isolation were to be used. Also, as purification / isolation is not mandatory, no loss of microorganisms during purification / isolation will take place. Also, in the case of samples containing potentially harmful microorganisms, the sample handling is safer as there are less sample preparation steps.

[0152] Even if purification / isolation of a sample is not mandatory in the calorimetric method of the present document, such purification / isolation may of course be performed if considered suitable.

[0153] It is often preferred to divide larger samples up into smaller pieces in order to ensure that the microorganisms present in the sample are dispersed in the metabolically differentiation inoculation media. The microbial sample is therefore typically homogenized or suspended in a liquid, such as phosphate-buffered saline, saline, or typical bacterial inoculation media, in order to break up the sample into smaller pieces and release microorganisms before being inoculated in the metabolically differentiating inoculation medium. The use of e.g , homogenization, shaking, vortexing etc. may be suitable means for dividing the sample up to disperse the microorganisms. Care must be taken so that the method used for dividing up / dispersing the microbial sample does not destroy the whole microbial community.

[0154] The present document also discloses a calorimetric method for the analysis of a microbial sample in order to determine if said sample is a) polymicrobial comprising a combination of at least two different types of microorganisms selected from the group consisting of Grampositive bacteria, Gram-negative bacteria and fungi, b) polymicrobial comprising a combination of at least two of the same types of microorganisms selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, or c) monomicrobial, said method comprising the steps of: a) providing a microbial sample; b) providing a set of metabolically differentiating inoculation media; c) inoculating and incubating said microbial sample in said metabolically differentiating inoculation media; d) obtaining calorimetric feature(s) from the inoculated and incubated sample of step c), thereby obtaining metabolic profiles from the different metabolically differentiating inoculation media; e) based on the metabolic profiles obtained in step d):

[0155] I) determining if said microbial sample is a) polymicrobial comprising a combination of at least two different types of microorganisms selected from the group consisting of Grampositive bacteria, Gram-negative bacteria and fungi, b) polymicrobial comprising a combination of at least two of the same types of microorganisms selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, or c) monomicrobial; and ii) optionally determining the taxonomic identity or of one or more of the microorganisms in said microbial sample, wherein said metabolically differentiating inoculation media allows for splitting up the metabolic profiles into the different types of microorganisms present in the microbial sample. The details of how to perform the steps of such as calorimetric method are as disclosed elsewhere herein.

[0156] The present document is also directed to a calorimetric method for the analysis of a microbial sample in order to determine if said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities, wherein said types of microorganisms are selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, said method comprising the steps of: a) inoculating and incubating said microbial sample in a set of metabolically differentiating inoculation media configured to split up a metabolic profile into the different types of microorganisms present in the microbial sample, wherein said set of metabolically differentiating inoculation media comprises a first metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-positive bacterium, a second metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-negative bacterium, and a third metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one fungus; b) obtaining calorimetric data from the inoculated and incubated sample of step a), thereby obtaining a metabolic profile from the first, second and / or third metabolically differentiating inoculation medium depending on what types of microorganisms are present in the microbial sample; c) based on the metabolic profiles obtained in step b), determining that said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms if a metabolic profile is obtained in at least two of said first, second and third metabolically differentiating inoculation medium; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if a metabolic profile is obtained in only one of said first, second and third metabolically differentiating inoculation medium.

[0157] Step b) of obtaining calorimetric data and step c) of determining of such a method correspond to steps d) of obtaining calorimetric data and step e) of determining as disclosed above. Further details regarding the microbial sample, media, analysis of calorimetric data etc. are as disclosed above.

[0158] The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.

[0159] EXPERIMENTAL SECTION

[0160] Materials and methods

[0161] Study setup overview

[0162] Non infected clinical tissue biopsies taken during orthopedic surgery of knee, hip, spine and shoulder (Karolinska Hospital, Sweden) were homogenized and used as clinical contrived samples for the microbial identification. The homogenized biopsies were combined with microbial growth media (see table 1) and spiked (microorganisms added) with mono or polymicrobial combinations at a 1 :1 ratio between the inoculated microorganisms. Further, in Example 5 (Fig. 6), a clinical sample with unknown microbial content was analysed using the calorimetric method of the present document.

[0163] General technical setup for generating contrived clinical samples

[0164] Clinical tissue biopsies from orthopedic surgery (Karolinska Hospital, Sweden) were homogenized and from the tissue homogenate, 50 pl were added to 250 pl of Mueller Hinton broth tryptic soy broth, brain-heart infusion, Sabouraud dextrose broth or yeast extract-peptone-dextrose broth, and thioglycolate enrichment broth (Sigma-Aldrich, Darmstadt, Germany), already dispensed in the sterile plastic insert in the microcalorimetry vial (Symcel AB, Soina, Sweden). The microcalorimetry vial was sealed and introduced in the calScreener (Symcel AB, Soina, Sweden). Samples were incubated and heat produced by each sample was measured for 24 h at 37°C. For the monomicrobial cultures 20 pl of the bacterial suspension was added to 300 pl of test media already in the titanium vials, for the polymicrobial cultures 10 pl of each microbial suspension was added to the test media. The technical setup was repeated in example 1 to 4 (Fig. 2-5), the variable in each example is the generation of differential samples with varying mono or polymicrobial organisms. In Example 5 (Fig. 6), a clinical polymicrobial biopsy sample was analysed using the method of the present document. No microorganisms were thus added to the biopsy of Example 5 (Fig. 6), but the microbial population analysed there was the population that was present in the infected biopsy taken from a patient. Mono and polymicrobial experimental setup

[0165] Example 1: Mixed microbial samples for polymicrobial analysis was made of i) Staphylococcus aureus + Pseudomonas aeruginosa and ii) Staphylococcus aureus + Candida albicans. See Fig. 2

[0166] Example 2: Mixed microbial samples for polymicrobial analysis was made of i) Staphylococcus aureus + Cory nebacteri urn striatum and for monomicrobial analysis ii) Staphylococcus aureus. See Fig. 3

[0167] Example 3: Mixed microbial samples for polymicrobial analysis was made of i) Pseudomonas aeruginosa + Escherichia coli and for monomicrobial analysis ii) Pseudomonas aeruginosa. See Fig. 4

[0168] Example 4: Mixed microbial samples for polymicrobial analysis was made of i) Candida glabrata + Candida albicans and ii) Candida albicans + Pseudomonas aeruginosa. See Fig. 5.

[0169] Example 5: Clinical microbial samples for polymicrobial analysis included Enterococcus faecalis and Klebsiella pneumoniae. See Fig. 6.

[0170] Results

[0171] Based on the calorimetric signal in different sets of the selective and / or differentiating media including MHB, FTM, Mannitol salt and Sabouraud G feature groups were defined and used for machine learning. 9 main parameter groups with the following features were used to build decision threes to make a prediction of the species identification:

[0172] 1 . Time to peak.

[0173] 2 Amplitude of highest peak with a length of > 1h and a prominence of > 1 % of the highest peak amplitude.

[0174] 3. Total area under the curve (sum of heat generated)

[0175] • Same as above but for 1st, 2nd, 3rd, 4thtime interval.

[0176] • Same as above but relative 1st, 2nd, 3rd, 4thtime interval

[0177] 4. Number of peaks with a prominence of at least 2% of the highest peak amplitude

[0178] • Same as above but for 1st, 2nd, 3rd, 4th, 5thtime interval.

[0179] • Same as above but for prominence 7,5%, 15%, 40%, 50% of highest peak amplitude and width 1h.

[0180] • Same as above but with width 2h, 4h, 8h.

[0181] 5. Exponential vs decay ratio

[0182] • From highest peak, descend left until 90% of peak value call this point (L_P90), keep descending until at 25% of peak value call this point (L_P25).

[0183] • From highest peak, descend right until 90% of peak value call this point (R_P90), keep descending until at 25% of peak value call this point (R_P25).

[0184] • Calculate L_AUC as area under curve from (L_P25) to (L_P90). • Calculate R_AUC as area under curve from (R_P25) to (R_P90).

[0185] • Calculate exponential vs decay ratio as (L_AUC) / (L_AUC+ R_AUC).

[0186] 6. Percentiles 5, 25, 50, 75, 95.

[0187] 7. Percentiles (85) - Percentiles (15).

[0188] 8. Arrays

[0189] • Multiple peaks with prominence 1 - 10 % of maximum peak value and width 2-10 h.

[0190] 9. Twin peaks detector

[0191] • Looks for two highest peaks above 5% prominence of the highest peak amplitude, that are not overlapping at least for 2h.

[0192] The combination of the outcomes from multiple decision trees is used for the identification of the individual subcomponents of an analyzed clinical sample.

[0193] Example 1 : Identification and separation of the microbial subcomponents of a microbial sample into polymicrobial and / or monomicrobial sample containing Gram-positive and Gram-negative bacteria or Gram-positive bacteria and fungi

[0194] The metabolic signal described in Figure 2A in MHB, FTM, Mannitol salt and Sabouraud G- media was analysed using the example decision tree of Figure 2B that is based on the following features: ) Twin peaks which are not overlapping: The presence of non-overlapping twin peaks with prominence > 5% of the highest peak amplitude in MHB media indicated that the sample is polymicrobial comprising of Gram-positive and Gram-negative microorganisms. ) Amplitude of highest peak: samples with peak prominence of > 15 % of highest peak amplitude in FTM media, indicated that the sample is polymicrobial and contains a Grampositive and Fungal microorganism. ) The sum of total heat generated in the Gram-positive selective media Mannitol salt with a time span 20 h allowed for Genus and Species identification of the individual Gram-positive subcomponents of polymicrobial sample. ) Arrays with prominence 1-10%: The presence of the peak arrays having a prominence 1-10% of the highest peak value and width a time span of 2-10h in Sabouraud G- media allows for Fungi species identification. ) Ratio above 0.22% of exponential vs decay signal in Sabouraud G- media: in combination with a comparison of the metabolic signal to a library of metabolic signals of different bacteria grown in the same conditions allows for Genus and Species identification of Gram- negative subcomponents of the polymicrobial sample.

[0195] Example 2: Identification and separation of the microbial subcomponents of a microbial sample to determine if it is polymicrobial with Gram-positive bacteria or monomicrobial with a Gram-positive bacterium

[0196] The metabolic signal described in Figure 3A was analysed using decision tree of Figure 3B based on following features obtained in MHB, FTM, Mannitol salt and Sabouraud G- media: ) Absence of metabolic activity in the selective Sabouraud G- media indicates that a polymicrobial sample can only be comprised of a combination of Gram-positive species or that the sample is comprised of a monomicrobial Gram-positive composition. ) The number of peaks > 2% of highest peak prominence in MHB media: indicated that it is a polymicrobial and not a monomicrobial Gram-positive sample. ) The ratio > 2% of exponential vs decay signal in FTM media: followed by in combination with a comparison of the metabolic signal to a library of metabolic signals of different bacteria grown in the same conditions allows for Genus and Species identification of Gram- negative subcomponents of the polymicrobial sample. ) The sum of total heat generated in the Gram-positive selective media Mannitol salt, allows for Genus and Species identification of the individual Gram-positive subcomponents of polymicrobial sample.

[0197] Example 3: Identification and separation of the microbial subcomponents of a microbial sample to determine if it is polymicrobial with Gram-negative bacteria or monomicrobial with a Gram-positive bacterium

[0198] The metabolic signal described in Figure 4A was analysed using example decision tree Figure 4B based on following features obtained in MHB, Mannitol salt and Sabouraud G- media:

[0199] 1) Absence of metabolic activity in the selective Mannitol salt media indicates that a polymicrobial sample can only be comprised of a combination of Gram-negative and Fungi or that the sample is comprised of a polymicrobial or monomicrobial Fungal composition.

[0200] 2) Ratio above 2% of exponential vs decay signal in MHB- media: in combination with a comparison of the metabolic signal to a library of metabolic signals of different bacteria grown in the same conditions allows for Genus and Species identification of Gram- negative monomicrobial sample.

[0201] 3) Twin peaks which are not overlapping: The presence of non-overlapping twin peaks with prominence > 25% of highest peak amplitude in MHB media indicated that the sample is polymicrobial comprising of Gram-negative and Gram-negative microorganisms.

[0202] 4) The sum of total heat generated in the Gram-negative selective media Sabouraud G- allows for Genus and Species identification of the individual Gramnegative subcomponents of polymicrobial sample. of the microbial of a clinical and / or monomicrobial containinq Gram-neqative bacteria and funqi and / or monomicrobial or with different

[0203] The metabolic signal described in Figure 5A was analysed using the example decision tree of Figure 5B based on following features obtained in MHB, FTM, Mannitol salt and Sabouraud G- media:

[0204] 1 . Absence of metabolic activity in the selective Mannitol salt media indicates that a polymicrobial sample can only be comprised of a combination of Gramnegative and Fungi or that the sample is comprised of a polymicrobial or monomicrobial Fungal composition.

[0205] 2. Total heat percentiles 5, 25, 50, 75, 90 in Sabouraud G- media. Comparison of total heat between a particular score and the scores of the rest of a group indicated that it is a polymicrobial sample comprising of Fungi and Fungi microorganisms.

[0206] 3. The ratio > 2% of exponential vs decay signal in FTM media: followed by comparison of the metabolic signal to a library of metabolic signals of different bacteria grown in the same conditions allows for Genus and Species identification of Fungi subcomponents of the polymicrobial sample

[0207] 4. The sum of total heat generated in FTM media allows for Genus and Species identification of the individual Fungi subcomponents of polymicrobial sample.

[0208] 5. Identification of non-overlapping twin peaks in MHB. Presence of nonoverlapping twin peaks with prominence >2% of highest peak amplitude indicated that the sample is polymicrobial comprising of Gram-negative and Fungi microorganisms.

[0209] 6. Identification of arrays with prominence 1 -10% of highest peak value in Sabouraud G- media. The presence of the peak arrays having a prominence 1- 10% of highest peak amplitude and width of 2-1 Oh allows for the identification of the Fungal species.

[0210] 7. The ratio >2% of exponential vs decay signal in Sabouraud G- media followed by comparison of the metabolic signal to a library of metabolic signals allows for for Fungi species identification. of the microbial clinical with unknown microbial content into its Gram-i and Gram-i monomicrobial com

[0211] The metabolic signal described in Figure 6A in MHB, FTM, Mannitol salt and Sabouraud G- media was analysed using the example decision tree of Figure 6B that is based on the following features: 1) Twin peaks which are not overlapping: The presence of non-overlapping twin peaks with prominence > 5% of the highest peak amplitude in MHB media indicated that the sample is polymicrobial comprising of Gram-positive and Gram-negative microorganisms.

[0212] 2) The sum of total heat generated in the Gram-positive selective media Mannitol salt: allowed for Genus and Species identification of the individual Gram-positive subcomponents of polymicrobial sample.

[0213] 3) Ratio above 2% of exponential vs decay signal in Sabouraud G- media: in combination with a comparison of the metabolic signal to a library of metabolic signals of different bacteria grown in the same conditions allows for Genus and Species identification of Gram- negative subcomponents of the polymicrobial sample.

[0214] Conclusion

[0215] Examples of the mono and polymicrobial bacterial communities in contrived tissue biopsies (Figs. 2-5) and a naturally infected microbial sample with unknown content (Fig. 6) combined with selected microbial growth media demonstrate the principle for identification and separation of a polymicrobial sample from a monomicrobial sample and how the subcomponents of individual spices in the samples can be identified using calorimetry. The identification of the microbial subcomponents is based on the methodology of i) permissive and nonpermissive growth media (i.e. metabolically differentiating media) and ii) combining defined feature groups from several different growth media to identify the specific microbial species that are present.

[0216] Polymicrobial with different types of microorganisms: The first hierarchy is illustrated in Fig. 1 step A, for inclusion or exclusion of a polymicrobial sample of mixed subcomponents. The use of permissive and nonpermissive growth media, allowed for the identification of a potential mixed polymicrobial sample, comprising of either Gramnegative, Gram-positive and / or fungi.

[0217] Polymicrobial with the same type of microorganisms: The second hierarchical level, Fig. 1 step B, allowed for the inclusion or exclusion of a polymicrobial sample consisting of more than one similar subcomponent of Gram-negative, Gram-positive or fungi simultaneously. Results were obtained using a combination of permissive and nonpermissive growth media and combining specifically defined feature groups for polymicrobial samples from the same microbial subcomponents. The output from steps one and two enabled determining if a sample was polymicrobial and if that sample is comprised of several simultaneous mixed subcomponents (Gram-negative, Grampositive and / or fungi) and / or if there is more than one subcomponent (several Gramnegative, Gram-positive and / or fungi) in the same sample (Poly similis). Species identification: The third hierarchical level is for the identification of the specific microbial subcomponents (species) that was described in Fig. 1 step C for a range of different species. The result was obtained by combining specifically defined feature groups in media selected for conferring a high identification resolution. Allowing for species identification both in monomicrobial samples, polymicrobial samples and poly similis samples.

[0218] In Example 5 (Fig. 6), it is demonstrated that the method of the present document allows the identification of the type and taxonomic identity of the individual microbial components of a clinical polymicrobial sample with unknown microbial content.

[0219] It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims

[0220] Unless expressly described to the contrary, each of the preferred features described herein can be used in combination with any and all of the other herein described preferred features.

[0221] REFERENCES

[0222] W02007 / 010379

[0223] Vazquez C. et al., Study of growth of Enterococcus faecalis, Escherichia coli and their mixtures by microcalorimetry, J Therm Anal Calorim, DOI 10.1007 / s10973-015-5203-y, 2016.

Claims

CLAIMS1 . A calorimetric method for the analysis of a microbial sample in order to determine if said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities, wherein said types of microorganisms are selected from the group consisting of Gram-positive bacteria, Gram-negative bacteria and fungi, said method comprising the steps of: a) providing a microbial sample; b) providing a set of metabolically differentiating inoculation media configured to split up a metabolic profile into the different types of microorganisms present in the microbial sample, wherein said set of metabolically differentiating inoculation media comprises a first metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gram-positive bacterium, a second metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one Gramnegative bacterium, and a third metabolically differentiating inoculation medium configured to provide a metabolic profile if said microbial sample comprises at least one fungus; c) inoculating and incubating said microbial sample in said first, second and third metabolically differentiating inoculation medium; d) obtaining calorimetric data from the inoculated and incubated sample of step c), thereby obtaining a metabolic profile from the first, second and / or third metabolically differentiating inoculation medium depending on what types of microorganisms are present in the microbial sample; e) based on the metabolic profiles obtained in step d), determining that said microbial sample is: i) a polymicrobial sample comprising at least two different types of microorganisms if a metabolic profile is obtained in at least two of said first, second and third metabolically differentiating inoculation medium; or ii) a monomicrobial sample or a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if a metabolic profile is obtained in only one of said first, second and third metabolically differentiating inoculation medium.

2. The calorimetric method according to claim 1 , wherein said first metabolically differentiating inoculation medium is selected from the group consisting of salt mannitol broth and CLED broth, said first metabolically differentiating inoculation medium beingsupplemented with at least one supplement selected from the group consisting of salts, such as sodium chloride and / or lithium chloride; antibiotics, such as colistin polymyxin B; and / or antimicrobials, such as phenyl ethanol.

3. The calorimetric method according to claim 1 or 2, wherein said second metabolically differentiating inoculation medium is selected from the group consisting of MacConkey broth, Hektoen enteric agar, and Sabouraud broth, said second metabolically differentiating inoculation medium being supplemented with at least one supplement selected from the group consisting of salts, such as eosin Y; sodium taurocholate; sodium selenite; antibiotics, such as vancomycin, cycloheximide, and / or azinomycin; antimicrobials such as brilliant green; and / or saccharides, such as lactose and / or glucose.

4. The calorimetric method according to any one of the preceding claims, wherein said third metabolically differentiating inoculation medium is selected from the group consisting of Sabouraud broth, malt extract broth, and yeast extract-peptone dextrose broth, said third metabolically differentiating inoculation medium being supplemented with at least one supplement selected from the group consisting of gentamycin, chloramphenicol, cephalothin, and cefamandole.

5. The calorimetric method according to any one of the preceding claims, wherein, if said microbial sample is determined to be a microbial sample according to step e)ii), said method comprises a further a step of determining that said microbial sample is: i) a polymicrobial sample comprising at least two different microorganisms of the same type but of different taxonomic identities if said metabolic profile(s) match(es) two or more monomicrobial metabolic profiles or matches a combined metabolic profile of a polymicrobial sample with known microbial content; or ii) a monomicrobial sample if the metabolic profile(s) match(es) a monomicrobial metabolic profile, wherein said metabolic profile(s) of said microbial sample are obtained from: a. at least one metabolically differentiating inoculation medium of the same type as the first, second or third metabolically differentiating inoculation medium in which a metabolic profile was obtained according to step e)ii) of any one of the preceding claims; and b. at least one non-selective general purpose inoculation medium, such as tryptic soy broth or Mueller Hinton broth.

6. The calorimetric method according to any one of the preceding claims, said method comprising a further step of determining the family, genus and / or species of said microorganisms by analysing said metabolic profiles.

7. The calorimetric method according to any one of the preceding claims said, wherein said method comprises a pre-incubation and re-loading step performed before incubating said microbial sample in said metabolically differentiating inoculation media, said pre- incubation and re-loading step comprising inoculating and incubating said microbialsample in at least one inoculation medium, calorimetrically following the metabolic activity of the incubated sample, and, when the R value for the metabolic rate is >0.98 for at least about 20 min or when the metabolic activity of the incubated sample reaches the inflection point or within 2 hours after said inflection point is reached, reloading one or more aliquot(s) of said incubated sample in said metabolically differentiating inoculation media before proceeding with the remaining steps of the method.

8. The calorimetric method according to any one of the preceding claims, wherein said microbial sample is a clinical sample from a subject, an environmental sample, a food or feed sample, and / or a purified microbiological sample, such as blood, tissue biopsy, synovial fluid, perinatal fluid, cerebrospinal fluid, pleural fluid, pericardial fluid, sonication fluid from an implant, urine, sputum, stool, and / or saliva.

9. The calorimetric method according to any one of the preceding claims, wherein said method is performed without first isolating said one or more microorganisms from said microbial sample.

10. The calorimetric method according to any one of the preceding claims, wherein said metabolic profiles are matched to metabolic profiles obtained from polymicrobial and / or monomicrobial samples with known microbial content.11 . The calorimetric method according to any one of the preceding claims, wherein the metabolic profiles are analysed using an identification algorithm matching a metabolic profile obtained according to any one of these claims against a database containing metabolic profiles of known microorganisms or wherein the metabolic profiles are analysed using a machine learning model trained on monomicrobial and polymicrobial samples.

12. A computer program comprising computer program code, the computer program code being adapted to, if executed on a processor, implement the step(s) of determining of the calorimetric method according to any one of the claims 1-11.

13. A computer program product comprising a computer readable storage medium, the computer readable storage medium having the computer program according to claim 12.

14. A computer-implemented method of determining the type and / or taxonomic identity of one or more microorganisms in a microbial sample according to any one of claims 1 to 11 , said computer-implemented method comprising the steps of: a) receiving calorimetric data from a calorimeter; b) using the calorimetric data to obtain calorimetric features; c) establishing a correlation between said calorimetric features and a plurality of microorganisms; determining the type and / or taxonomic identity based on said calorimetric features.

15. A computer implemented method for training a machine learning model to identify one or more microorganisms, the method comprising:-obtaining calorimetric feature data, the calorimetric feature data being derived from calorimetric data being indicative of the metabolic activity of the one or moremicroorganism(s) in a set of different metabolically differentiating inoculation media as defined in any one of claims 1-5;-obtaining media data indicative of the set of different metabolically differentiating inoculation media; -obtaining microorganism data indicative of a set of microorganisms from which the one or more microorganisms are identified;-generating a trained machine learning model by training, using training data comprising the calorimetric feature data, the media data and the microorganism data, a machine learning model to output, based on input operating data, identification of one or more microorganisms.

16. A kit for performing a calorimetric method according to any one of claims 1-11 , said kit comprising or consisting of a) a set of metabolically differentiating inoculation media; b) a training data set which contains metabolic profiles of known microorganisms incubated in the same metabolically differentiating inoculation media; c) a software component comprising a function that determines the type and / or taxonomic identity of a microorganism.

Images