Methods for detecting apoptosis

Abstract

The present invention concerns a method for detecting apoptosis, wherein the method comprises the steps of: heating a non-purified test sample comprising a population of cells; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level. Further disclosed is a method for identifying an apoptosis-inducing agent, a method of investigating apoptotic processes in a population of cells and a method of diagnosing an apoptotic-associated condition or disease in a subject, all using the above method.

Description

The present invention relates to methods for detecting apoptosis. Apoptosis is a repertoire of controlled cellular processes that eventually lead to cell death and systematic degradation of cellular components. Apoptosis is a form of programmed cell death in which cells are degraded in an ordered form. Apoptosis inducing pathway are typically divided into extrinsic and intrinsic pathways. The textbook scenario is that extrinsic pathways are induced by extracellular signals leading to intracellular activation of caspase 8, an initiator caspase, whereas apoptosis proceeding through the intrinsic pathway is typical induced by some type of cellular stress, leading to activation of initiator caspase 9. The initiator caspases subsequently cleave and activate effector caspases such as caspase 3 and 7. Apoptosis plays important roles in a range of pathological conditions including neurodegenerative, immune suppressant, liver and infection diseases. Apoptosis is also an important component in the action of some cytotoxic drugs such as anti-cancer or immune suppressant drugs. Thus, activation of cellular pathways that block apoptosis is a means for cancer cells to establish resistance against anti-cancer drugs. In neurological diseases such as Alzheimer’s or Parkinson diseases, the attenuation of apoptotic processes by drugs is desirable to slow disease progression. Therefore stringent methods to detect apoptosis are valuable as research tools and in clinical assays. Typically, the detection of apoptosis within a cell is performed with imaging or cell sorting techniques, which are time consuming and / or do not yield details about the apoptosis process and / or do not adequately detect apoptosis during the earliest stages of the apoptotic process. Examples of assays used to detect specific components of apoptosis include the annexin 5 assay in which distortions of lipid organisation on the cell membrane induced by apoptosis is detected, and assays detecting caspase activity or the cleavage of caspase targets. These assays are however not always applicable and can be prone to experimental errors. There remains a need in the field for further and improved methods of detecting apoptosis. Particularly advantageous would be methods for detecting apoptosis earlier in the apoptotic process than can be detected using imaging or sorting techniques. There remains a need for methods that can resolve the apoptosis process in greater detail. The present inventors have identified an advantageous method for detecting apoptosis, which advantageously permits detection soon after the commencement of apoptosis. Thermal shift assays have been developed in the art which can assess protein-ligand binding where the protein is in purified form. These assays have been developed on the basis of two principles, namely that a purified protein will melt and unfold at a particular temperature and that the binding of a ligand to a protein will thermally stabilise the protein. Thus, the binding of a ligand to a protein can be detected on the basis that the purified protein will show an increase in thermal stability once a ligand is bound and hence the protein will melt at a higher temperature once ligand is bound than purified protein alone. The methods of the invention comprise the use of Cellular Thermal Shift Assay (CETSA), which is described in WO 2012 / 143714A1. Prior to the invention of CETSA, thermal shift assays known in the field were only used in connection with purified protein. WO2012 / 143714 describes the use of CETSA for detecting stability changes of target proteins within or on intact cells following drug-target protein binding, in non-purified samples, i.e. without purification of the target protein. CETSA allows for global studies of protein stability changes in intact cells, primarily reflecting interaction changes for proteins. Most applications of CETSA have so far focused on identifying the direct binding of drugs with their cognate protein target(s), including the use of proteome-wide mass-spectrometry CETSA (MS-CETSA) for discovering new target proteins. At the same time, multiple explorative studies have established that CETSA can yield distinct information on the interaction changes between proteins with other cellular molecules. However, for a CETSA shift to be measurable, interaction / biochemical changes are required at relatively high cellular stoichiometries, with protein stability significantly affected by the interaction(s). This presents challenges and an expected limitation on the utility of the method in analyzing less dominant intracellular effects, particularly in complex, non-purified samples. In one aspect, the present invention provides a method for detecting apoptosis, wherein the method comprises the steps of: heating a non-purified test sample comprising a population of cells; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, and / or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level, and wherein said biomarker protein is selected from the group consisting of ADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CNOT2, COA7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPAO, HNRNPA3, HNRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, N0P14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8. Alternatively viewed, the method comprises determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof as compared to a control level, and / or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample as compared to a control level, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to the control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to the control level. Alternatively viewed, the method comprises determining the soluble and / or insoluble level of a biomarker protein in said heated test sample or in a protein fraction thereof, wherein apoptosis is detected by an increase in the soluble level of the biomarker protein as compared to a control level, and / or by a decrease in the insoluble level of the biomarker protein as compared to a control level. Alternatively viewed, the method comprises determining the soluble and / or insoluble level of a biomarker protein in said heated test sample or in a protein fraction thereof as compared to a control level, wherein apoptosis is detected by an increase in the soluble level of the biomarker protein as compared to the control level, and / or by a decrease in the insoluble level of the biomarker protein as compared to the control level. In all aspects and embodiments described herein, reference to the level of a protein is a reference to the level of a biomarker protein as defined herein. In all aspects and embodiments described herein, reference to determining or comparing a “level” of a protein, includes determining or comparing the levels of a group or panel of proteins, meaning the individual level of each protein in the group or panel, or the levels of the group or panel in a summed or averaged manner. The level determined or compared may be the soluble and / or the insoluble level, unless it is clearly stated or it is clear from the technical context that the soluble level in particular, or the insoluble level in particular, is required. Apoptosis is detected by a difference between the level of the soluble and / or insoluble biomarker protein as compared to a control level. Specifically, the difference is indicative of apoptosis when a) the difference is an increase in the level of the soluble biomarker protein as compared to a control level (i.e. a control level of the soluble biomarker protein), or b) the difference is a decrease in the level of the insoluble biomarker protein as compared to a control level (i.e. a control level of the insoluble biomarker protein). The biomarker proteins of the present invention are more thermally stabile (more soluble) under apoptotic conditions than under non-apoptotic conditions. Thus in the CETSA-based methods of the invention, if the soluble form of the biomarker proteins is detected at an increased level in the test sample as compared to in the control sample, then apoptosis is detected. Conversely, in the CETSA-based methods of the invention, if the insoluble form of the biomarker proteins is detected to a decreased level in the test sample than in the control sample, then apoptosis is detected. The term “difference” in the level of a biomarker protein vs. control is used herein, however, in all aspects and embodiments, it is expressly meant that for detection of apoptosis, the difference is an increase vs. control in the context of soluble biomarker protein, and / or a decrease vs. control in the context of insoluble biomarker protein. The methods of the invention for detecting apoptosis can be viewed as methods for determining the apoptotic status (or apoptotic index) of a population of cells. In such methods, the apoptotic status or index of the population of cells is determined by detecting a difference between the determined level of at least one of the soluble or insoluble biomarker proteins as compared to a control level. The extent of apoptotic status (or index) may correlate with the number of biomarker proteins for which there is detected a difference between the determined level of the soluble and / or insoluble biomarker protein as compared to a control level. Alternatively, the extent of apoptotic status (or index) may correlate with the extent (magnitude) of the difference between the determined level(s) of the soluble and / or insoluble biomarker protein(s) as compared to control level(s). The extent of the difference may be determined on an individual protein basis, or on the basis of the sum of the differences in the levels of multiple soluble or insoluble biomarker proteins as compared to control levels. Thus, there is also provided a method for determining the apoptotic status (or “apoptotic index”) of a population of cells, wherein the method comprises the steps of: heating a non-purified test sample comprising said population of cells; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, and / or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, wherein the apoptotic status of the population of cells is determined as a function of the magnitude of an increase in the level of the soluble biomarker protein as compared to a control level, and / or as a function of the magnitude of a decrease in the level of the insoluble biomarker protein as compared to a control level, and / or wherein the apoptotic status of the population of cells is determined as a function of the number of biomarker proteins for which an increase in the level as compared to a control level is determined, and / or by the number of biomarker proteins for which a decrease in the level of the biomarker protein as compared to a control level is determined, and wherein said biomarker protein is selected from the group consisting of ADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CNOT2, COA7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPAO, HNRNPA3, HNRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, NOP 14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8. By “determined as a function of’ is meant correlates to, preferably positively correlates to. Alternatively viewed, there is invention provided a method of detecting or determining the enhancement of apoptosis in a cell population, which is as described above, and wherein the enhancement of apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level; or wherein the enhancement of apoptosis is determined as a function of the magnitude of an increase in the level of the soluble biomarker protein as compared to a control level, and / or as a function of the magnitude of a decrease in the level of the insoluble biomarker protein as compared to a control level, and / or wherein the enhancement of apoptosis is determined as a function of the number of biomarker proteins for which an increase in the level as compared to a control level is determined, and / or by the number of biomarker proteins for which a decrease in the level of the biomarker protein as compared to a control level is determined, The extent of the difference (i.e. increase of soluble protein or decrease of insoluble protein) may be determined on an individual biomarker protein basis, or on the basis of the sum of the differences in the levels of multiple soluble or insoluble biomarker proteins as compared to control levels. In all aspects and embodiments, analysis of the soluble levels of biomarker protein(s) is preferred. As used herein, “apoptosis” refers to cell death by an intracellular controlled process. Apoptosis is characterized by a condensation and subsequent fragmentation of the cell nucleus, and systematic degradation of cellular components, during which the plasma membrane remains intact. The apoptosis detected according to the present invention may be any category or type of apoptosis, including but not limited to extrinsic apoptosis, intrinsic apoptosis, caspase-mediated apoptosis, caspase-independent apoptosis, tumour necrosis factor (TNF)-related apoptosis, FAS ligand-mediated apoptosis, nuclear apoptosis, early nuclear apoptosis, drug-induced apoptosis, pathological apoptosis, aberrant apoptosis, and so on. The apoptosis detected is preferably that induced by a test condition (discussed elsewhere herein) to which the test sample has been (or is) exposed. For instance, the apoptosis detected may be that induced by a test molecule (discussed elsewhere herein, e.g. a drug) to which the test sample has been (or is) exposed. Preferably the apoptosis is nuclear apoptosis (which may be drug-induced), more preferably early nuclear apoptosis (which may be drug-induced). As is clear, the apoptosis detected or investigated according to the present invention is apoptosis of the cells in the non-purified test sample. A “biomarker protein”, as the term is used herein, is a protein that is within or on a cell. The cell may be any cell, including a prokaryotic or eukaryotic cell. Preferably the cell is a mammalian cell, more preferably a human cell e.g. a cell from a human patient. The cell may be an animal cell, particularly a cell from an animal patient, or a disease model, e.g. dog, cat, monkey, rabbit, mouse, rat, etc.). Thus, preferably the biomarker protein is a mammalian protein, more preferably a human protein. The cell may be a plant cell, thus the biomarker protein may be a plant protein. The cell may be a bacterial cell, thus the biomarker protein may be a bacterial protein. The biomarker proteins named herein are human proteins, but the biomarker proteins may be homologous proteins from other species. A protein “within” a cell is a protein located on the interior of the cell, i.e. inside the space defined by the cell membrane, for instance but not limited to in the cytosol or nucleus. A protein “on” a cell is a protein located accessible from outside the cell, i.e. from the exterior of the cell membrane, and may be tethered or otherwise anchored to the cell membrane. Proteins 5 located in the cell membrane and on only the internal side of the cell membrane would be considered to be “within the cell”. Proteins spanning the cell membrane may be considered to be proteins both “within” and “on” the cell. However, these distinctions are not relevant to the present methods. Extracellular proteins, i.e. those not within or on the cell, are not proteins of interest according to the present invention. 10 The biomarker proteins of the present invention have been demonstrated to be indicative of apoptosis when observed to “shift” after being heated in a CETSA assay (i.e. when a change in their soluble and / or insoluble level as compared to a control sample is observed). The biomarker is a, i.e. at least one, i.e. one or more of the following proteins: Biomarker Protein Name Uniprot Accession No. ADNP Activity-dependent neuroprotector homeobox protein Q9H2P0 AIFM1 Apoptosis-inducing factor 1 095831 BCAP31 B-cell receptor-associated protein 31 P51572 BCLAF1 Bcl-2-associated transcription factor 1 Q9NYF8 CBX1 Chromobox protein homolog 1 P83916 CBX5 Chromobox protein homolog 5 P45973 X—'- JL / / Cilia- and flagella-associated protein 100 Q494V2 CDC5L Cell division cycle 5-like protein Q99459 CHCHD3 MICOS complex subunit MIC 19 Q9NX63 CHD4 Chromodomain-helicase-DNA-binding protein 4 Q14839 CNOT2 CCR4-NOT transcription complex subunit 2 Q9NZN8 COA7 Cytochrome c oxidase assembly factor 7 Q96BR5 DDX18 ATP-dependent RNA helicase DDX18 Q9NVP1 DDX21 ATP-dependent RNA helicase DDX21 Q9NR30 ESF1 ESF1 homolog Q9H501 FTSJ3 pre-rRNA 2'-O-ribose RNA methyltransferase Q8IY81 GABPB1 GA-binding protein subunit beta-1 Q06547 GNL2 Nucleolar GTP-binding protein 2 Q13823 HNRNPAO Heterogeneous nuclear ribonucleoprotein AO Q13151 HNRNPA3 Heterogeneous nuclear ribonucleoprotein A3 P51991 HNRNPC Heterogeneous nuclear ribonucleoproteins C1 / C2 P07910 HNRNPU Heterogeneous nuclear ribonucleoprotein U Q00839 IMMT MICOS complex subunit MIC60 Q16891 LMNA Prelamin-A / C P02545 LMNB1 Lamin-B 1 P20700 LMNB2 Lamin-B2 Q03252 LUC7L3 Luc7-like protein 3 095232 MATR3 Matri n-3 P43243 ME2 Malic enzyme 2 P35227 ME3 Malic enzyme 3 QI6798 MORC3 MORC family CW-type zinc finger protein 3; QI6798 NCL Nucleolin P19338 NOP 14 Nucleolar protein 14 P78316 PARP1 Poly [ADP-ribose] polymerase 1 P09874 PPM1G Protein phosphatase, Mg2+ / Mn2+ dependent 1G 015355 PRKDC DNA-dependent protein kinase catalytic subunit; P78527 RBM5 RNA-binding protein 5 P52756 RBMX RNA-binding protein X P38159 RRP1 Ribosomal RNA processing protein 1 P56182 RSF1 Remodeling and spacing factor 1 Q96T23 SMARCC1 SWI / SNF complex subunit SMARCC1 Q92922 TFAM Transcription factor A Q00059 TOPI DNA topoisomerase 1 P11387 U2SURP U2 snRNP-associated SURP motif-containing protein 015042 UBR7 Putative E3 ubiquitin-protein ligase UBR7 Q8N806 UBTF Nucleolar transcription factor 1 Pl 7480 WDHD1 WD repeat and HMG-box DNA-binding protein 1 075717 XRCC5 X-ray repair cross-complementing protein 5 P13010 XRCC6 X-ray repair cross-complementing protein 6 P12956 ZC3H13 Zinc finger CCCH domain-containing protein 13 Q5T200 ZMYND8 Protein kinase C-binding protein 1 Q9ULU4 In the methods of the invention, apoptosis may be determined by an increase in the level of a soluble or a decrease in the level of an insoluble biomarker protein as compared to a control level. Preferably, apoptosis is detected by a difference between the soluble and / or insoluble levels of a plurality of biomarker proteins in the heated test sample (or protein fraction thereof) as compared to the control levels of said biomarker proteins. For each biomarker protein, the control level of soluble / insoluble protein is the level of the same biomarker protein in a control sample (discussed in more detail below). The combined (e.g. summed or averaged) levels of the soluble biomarker proteins can be compared to the combined (e.g. summed or averaged) control levels thereof. Likewise, the combined (e.g. summed or averaged) levels of the insoluble biomarker proteins can be compared to the combined (e.g. summed or averaged) control levels thereof. The methods of the invention may comprise determining the soluble and / or insoluble levels of a plurality of biomarker proteins, e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 17, at least 20, at least 25, at least 30, at least 34, at least, 35, at least 38, at least 50 or all of said biomarker proteins. The methods of the invention may comprise determining the soluble and / or insoluble levels of at least 20%, preferably at least 25%, at least one third, at least 50%, at least two thirds, or at least 75%, more preferably all of said biomarker proteins. When such a plurality of biomarker proteins is assessed, apoptosis may be detected by determining a difference between the soluble and / or insoluble level of at least one of the biomarker proteins as compared to the control level thereof. Preferably, apoptosis is detected by determining a difference between the soluble and / or insoluble level of at least 25%, more preferably at least one third, more preferably at least 50%, more preferably at least two thirds, more preferably at least 75%, more preferably all of the biomarker proteins assessed, as compared to the control level of the corresponding biomarker protein. The biomarker proteins described herein need not all be assessed, and nor must soluble and / or insoluble levels of all of them be determined to be different as compared to the control level, in order for apoptosis to be detected. The biomarker proteins are a useful panel of biomarkers for apoptosis, wherein one or more, but typically a plurality of the biomarker proteins, will be observed to “shift” in CETSA assays of cells that are undergoing an apoptotic process. Precisely which of the biomarker proteins will “shift” in any given apoptotic scenario is not important for detection of apoptosis; it is sufficient for any one or more, preferably a plurality, of the biomarker proteins to be determined at a different soluble and / or insoluble level as compared to the control level. Different subsets of the biomarker proteins listed above may be particularly indicative of apoptosis in different apoptotic scenarios. Thus, apoptosis may be detected by a difference between the soluble and / or insoluble level of at least 2, more preferably at least 3, more preferably at least 4, more preferably at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 17, at least 20, at least 25, at least 30, at least 34, at least, 35, at least 38, or each of said biomarker proteins as compared to the control level of the corresponding biomarker protein. The methods of the invention may comprise determining the soluble and / or insoluble levels of substantially all, preferably all, proteins in the test sample or aliquot or protein fraction thereof being analysed. This is possible using, for instance, mass spectrometry detection techniques. Throughout this application, reference to a biomarker protein is intended to refer also to a plurality of biomarker proteins. In all cases, the biomarker protein is preferably not directly bound by any test molecule (e.g. drug) that is applied to the sample, or administered to the subject from which the sample is obtained. Thus preferably the biomarker protein is not a target protein for a drug that is applied to the sample, or administered to the subject from which the sample is obtained. Preferably, the biomarker protein is selected from the group consisting of CDC5L, CHD4, DDX21, FTSJ3, HNRNPA0, HNRNPA3, LMNB1, LMNB2, MATR3, M0RC3, NOP14, PARP1, RBMX, SMARCC1, U2SURP, XRCC5, XRCC6, ZC3H13, and ZMYND8. The methods of the invention may comprise determining the level of one or more, preferably a plurality of biomarker proteins, e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, or all of said preferred biomarker proteins. The levels of further biomarker proteins may also be determined. Apoptosis may be detected by a difference between the level of at least one, preferably between the level of a plurality e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, or each of said biomarker proteins as compared to the control level of the corresponding biomarker protein. The methods of the invention may comprise determining the levels of at least 20%, preferably at least 25%, at least one third, at least 50%, at least two thirds, or at least 75%, more preferably all of said biomarker proteins. Apoptosis may be detected by determining a difference between the level of at least 25%, more preferably at least one third, more preferably at least 50%, more preferably at least two thirds, more preferably at least 75%, more preferably all of the biomarker proteins assessed, as compared to the control level of said biomarker protein. Preferably, the biomarker protein is selected from the group consisting of CDC5L, CHD4, FTSJ3, HNRNPAO, HNRNPA3, MORC3, NOP14, RBMX, SMARCC1, U2SURP, XRCC5, XRCC6, ZC3H13, and ZMYND8. The methods of the invention may comprise determining the level of one or more, preferably a plurality of biomarker proteins, e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, or all of said preferred biomarker proteins. The levels of further biomarker proteins may also be determined. Apoptosis may be detected by a difference between the level of at least one, preferably between the level of a plurality e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, or each of said biomarker proteins as compared to the control level of the corresponding biomarker protein. The methods of the invention may comprise determining the levels of at least 20%, preferably at least 25%, at least one third, at least 50%, at least two thirds, or at least 75%, more preferably all of said biomarker proteins. Apoptosis may be detected by determining a difference between the level of at least 25%, more preferably at least one third, more preferably at least 50%, more preferably at least two thirds, more preferably at least 75%, more preferably all of the biomarker proteins assessed, as compared to the control level of said biomarker protein. Alternatively, the biomarker protein may be selected from the group / panel consisting of BCAP31, COA7, CCDC37, DDX21, LMNB1, LMNB2, MATR3, ME2, ME3, PARP1, PRKDC, TFAM, U2SURP, XRCC5, XRCC6, ZMYND8. With these biomarker proteins, apoptosis may be advantageously detected in extremely short time-frames after the induction of apoptosis (e.g. within 1 hour of apoptosis). The methods of the invention may comprise determining the level of one or more, preferably a plurality of biomarker proteins, e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15 or all of this group of biomarker proteins. The levels of further biomarker proteins may also be determined. Apoptosis may be detected by a difference between the level of at least one, preferably between the level of a plurality e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15 or each of said biomarker proteins as compared to the control level of the corresponding biomarker protein. The methods of the invention may comprise determining the levels of at least 20%, preferably at least 25%, at least one third, at least 50%, at least two thirds, or at least 75%, more preferably all of said biomarker proteins. Apoptosis may be detected by determining a difference between the level of at least 25%, more preferably at least one third, more preferably at least 50%, more preferably at least two thirds, more preferably at least 75%, more preferably all of the biomarker proteins assessed, as compared to the control level of said biomarker protein. The biomarker protein may be selected from the group / panel consisting of ADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CDC5L, CHCHD3, CHD4, CNOT2, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPA0, HNRNPA3, FINRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, M0RC3, NCL, NOP14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13 and ZMYND8. The methods of the invention may comprise determining the level of one or more, preferably a plurality of e.g. at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45 or all of this group of biomarker proteins. The levels of further biomarker proteins may also be determined. The methods of the invention may comprise determining the levels of at least 20%, preferably at least 25%, at least one third, at least 50%, at least two thirds, or at least 75%, more preferably all of said biomarker proteins. Apoptosis may be detected by determining a difference between the level of at least 25%, more preferably at least one third, more preferably at least 50%, more preferably at least two thirds, more preferably at least 75%, more preferably all of the biomarker proteins assessed, as compared to the control level of said biomarker protein. Apoptosis may be detected by a difference between the level of at least 2, more preferably at least 3, more preferably at least 4, more preferably at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45 or each of said biomarker proteins as compared to the control level of the corresponding biomarker protein. When considering any group of biomarker proteins described herein, preferred methods include determining the soluble and / or insoluble level of at least two, more preferably at least three, at least four or at least five, of said biomarker proteins, and wherein apoptosis is detected by an increase in the level of soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level, for each of said biomarker proteins. When considering any group of biomarker proteins described herein, preferred methods include determining the soluble and / or insoluble level of at least eight, more preferably at least 10, at least 16 at least 20 of said biomarker proteins, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level for at least half of said biomarker proteins. As used herein, the term “test sample” is used to indicate a sample on which the steps of a method of the invention are performed. The steps of a method of the invention may also be performed on a “control sample”, however, this is not essential. The methods of the invention may comprise the comparison of data obtained using a test sample with control data (i.e. data obtained using a control sample) however, such control data may have been obtained previously. In other words, the data obtained with the control sample may be in the form of reference data, e.g. a reference database, and the methods of the invention need not comprise the performance of steps on a control sample. Data in this regard is preferably the level of one or more biomarker proteins, as described anywhere elsewhere herein. The biomarker protein(s) is / are within or on a cell, and the non-purified test sample used in the invention comprises said cell, e.g. comprises a plurality, i.e. population, of said cells (i.e. a “cell population”). Thus, the sample preferably comprises the preferred cells described above. The invention is concerned with analysis of impure samples. The methods of the invention are performed on a “non-purified” test sample, i.e. a crude or complex sample. Control samples as referred to herein are also non-purified, and the discussion herein in relation to test samples applies equally to control samples. As used herein, this means that the method is performed without the biomarker protein(s) having been purified. The sample is thus one that has not been subjected to a purification process that would result in the purification of the biomarker protein(s). For instance, the sample has not been subjected to steps of extraction, precipitation and separation, e.g. by centrifugation or chromatography in order to purify the biomarker protein(s). The term “non-purified sample” means that the biomarker protein(s) is / are present with other proteins and biomolecules in the sample. The term does not however exclude purification at the cellular level, i.e. separation or removal of a particular cell type or cell types from within the sample prior to the performance of the invention. Prior to the heating step (step (a)), discussed below, non-purified samples may be prepared and / or processed. Such preliminary steps may comprise seeding cells in suitable media, treating cells with drugs or test molecules, incubating, washing, detaching, pelleting cells, re-suspending, etc. Such steps may also be comprised in the methods of the invention. None of these steps constitutes purifying the sample, because the biomarker protein(s) is / are still present in a non-purified form in or on the cells. Likewise, in the present methods, it is permitted to perform a step of cell resolution prior to any step of determining the level of a protein. Such a cell resolution step may allow the analysis of the levels of protein in particular cell types which have been isolated from a heterogenous sample (e.g. a tumour). In this respect, the invention provides a further step of cell sorting the samples, before or after the initial heating step. If the sample is a tumour sample, a step of tumour matrix degradation may also be necessary prior to any step of cell sorting. Methods for sorting cells are well known in the art e.g. FACS. Thus, the methods of the invention optionally comprise a step, prior to the protein level determination step, and before or after the heating step, of cell sorting to isolate a cell type of interest from the sample. Cell selection, resolution, etc., e.g. cell sorting, results in a new non-purified sample for use in the steps of the method of the invention; the sample is still non-purified, as the term is used herein, because the biomarker protein(s) has / have not been purified, i.e. is / are still present with other proteins and biomolecules in the sample. A non-purified sample may however comprise a heterogenous cell population. The non-purified sample preferably comprises intact cells. However, the non-purified sample may be a whole cell extract or whole cell lysate. The non-purified sample, may be a clinical or environmental sample. The non-purified sample may be obtained directly from a human or animal subject (e.g. patient). The non-purified sample may be a tissue sample (e.g. epithelium, muscle, nervous or connective tissue), blood sample, serum sample, plasma sample, lymph sample, cerebrospinal fluid sample, mucus sample, urine sample, or faeces sample, etc. The sample may be a liquid or solid tumour sample. The sample may also be obtained from a plant or part thereof, e.g. may be a plant extract. The non-purified sample may also be a cell colony or cell culture. A cell colony is a circumscribed group of cells, normally derived from a single cell or small cluster of cells growing on a solid or semi-solid medium (e.g. culture media with the addition of 0.1% or greater agar). The non-purified sample may be a liquid culture of cells. A liquid culture of cells may comprise cells which have all originated from a single cell i.e. the cells within the liquid culture may be clonal, or the liquid culture may comprise a suspension of different cells. The cells of the colonies or in liquid culture may be prokaryotic i.e. bacteria or eukaryotic cells e.g. yeast, unicellular eukaryotes such as Leishmainia, insect cells or mammalian cells or cell lines. Cells in liquid culture or grown as colonies may be formed as E.coli, Bacillus suhtilis, Streptococcus lactis, Streptococcus lividens, Lactococcus lactis, Staphylococcus aureas, Aspergillus niger, Picia pastoris, Saccharomyces cerevisiae or Schizosaccaromyces pombe. The methods of the invention comprise heating the non-purified test sample. This produces a heated non-purified test sample (referred to herein simply as a heated test sample). Applying a temperature to a sample will cause melting / unfolding and subsequently precipitation of proteins that are thermally unstable at that temperature. CETSA assays are predicated on thermally unstable proteins being precipitated by the heating step, whereas proteins that are thermally stable at the heating temperature applied remain properly folded and so soluble. Hence, the solubility of a protein after the heating step is linked to its thermal stability. Thus, the detection of a soluble protein (e.g. in a soluble protein fraction) after heat treatment indicates that it is properly folded and thermally stabile at the heating temperature. The present inventors have surprisingly determined that CETSA can also be used to detect apoptosis in (or provide information on the apoptotic state of) a population of cells by observing changes in the thermal stability of one or more identified CETSA biomarkers of apoptosis between test and control samples. The present inventors have demonstrated that this is possible even in non-purified samples, which are complex and have a high level of background “noise”. A sample that has been heated according to the present methods will thus comprise a mixture of proteins that are soluble (thermally stable at the temperature of heating) or insoluble (denatured by the temperature of heating). Thus, a "soluble” or “stable” protein can be defined in reference to possession of a native or native-like conformation, or as one which remains soluble (e.g. remains in the supernatant after centrifugation of a heated non-purified sample). Likewise, an "insoluble" or “destabilized” protein can be defined in reference to a loss of a native or native-like conformation, or as one which is not soluble (e.g. is not in the supernatant, but rather is in the insoluble protein fraction after centrifugation of a heated non-purified sample). It is possible that a protein may unfold and precipitate at one specific temperature. It is also possible that a protein may precipitate over a small temperature range. The “initial melting temperature” is the first temperature in the range at which the protein precipitates, i.e. the first temperature at which the protein is detected in the insoluble fraction, and the temperature at which the first loss of soluble protein is detected (e.g. is detected in the soluble fraction). Thus, the initial melting temperature is the lowest temperature at which a protein begins to precipitate e.g. at least 5%, or at least 10%, of the protein is precipitated. Conversely, the final melting temperature is the first temperature at which essentially no soluble protein of interest is detected (e.g. is detected in the soluble fraction.) e.g. less than 5% of the protein is in soluble form. Therefore, when a protein precipitates over a temperature range, the protein may begin to precipitate or unfold at a particular temperature at which point the amount of soluble protein present will begin to decrease and the amount of insoluble protein present will increase (since thermal stability is linked to solubility). Therefore, some soluble protein may still be detectable at the initial melting temperature until a slightly higher temperature is applied, at which point little or no soluble protein is detectable. The temperature which can be applied in the present invention may be any temperature sufficient to cause the precipitation of the biomarker protein of interest. The heating step may comprise heating the sample to a single temperature. The single temperature should be at or above the initial melting temperature of the biomarker protein. The initial melting temperature of the protein may be known. Even if the initial melting temperature of the protein is not known, the method may still comprise heating the sample to a single temperature (or at least one temperature) at or above the initial melting temperature of the protein. A suitable temperature can be determined via routine optimisation. If after heating at a particular temperature no biomarker protein of interest is identified in the insoluble protein fraction, or no loss of the biomarker protein of interest is identified in the soluble fraction, then the initial melting temperature of the biomarker protein has not been reached. Mammalian proteins may begin to denature at 37°C, and a large majority will be expected to have begun to denature by 60°C. Therefore, the sample is preferably heated to a temperature above 37°C, preferably above 40°C, e.g. above 50°C. A higher or lower temperature may however be used if the biomarker protein of interest is not a mammalian protein, e.g. a higher temperature may need to be used if the protein is from a thermophilic bacterium. The skilled person will readily be able to determine suitable heating temperatures for their desired purposes. The methods of the invention may comprise repeated performance of the steps of the method using one or more different temperatures in said step of heating, wherein apoptosis is detected by a difference between the level of the soluble or insoluble biomarker protein as compared to a control level, after heating to at least one temperature. Preferably it is desirable to expose the test sample to a series of temperatures over a temperature range. A series of temperatures refers to at least two different temperature points, preferably at least three, four, five or six temperature points. A temperature range is the temperature range over which the biomarker protein of interest unfolds and precipitates. The temperature range should include at least one temperature at or above the initial melting temperature of the biomarker protein of interest, which is as discussed above. The temperature points in the series of temperatures may differ from the adjacent temperature points by about 2, 3, 4, 5, 6, 7, 8, 9 or 10 °C, and the difference between temperature points need not be constant. In preferred methods, the heating step comprises heating the non-purified sample to a series of temperatures in the temperature range of 30 to 80°C, more preferably 35 to 70°C, more preferably 37°C to 60°C, 37°C to 59°C, or 37°C to 58°C, e.g. 40°C to 55°C. Measurements may be taken at any interval, but typically at one degree increase, two degree increase or three degrees increase in temperature. The sample may be heated to any temperature, e.g. at any one of more of 37, 38 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58 and 59°C„ as long as one of the temperatures is equal to or higher than the initial melting temperature for the biomarker protein of interest. Temperatures below 37°C and above 59°C may also be used when determined suitable by the user of the method; as mentioned above, this is within the general competencies of the person of ordinary skill in the art. Preferred temperature series comprise heating to temperatures of i) 37°C, 47°C, 50°C, 52°C, 54°C and 57°C; ii) 37°C, 47°C, 50°C, 53°C, 55°C and 59°C; or iii) 37, 42, 46, 52 and 58°C. Where the sample is heated over a temperature range, this can be carried out in a PCR machine where an initial temperature can be set and then increased by the desired amount after a particular amount of time e.g. 1, 2, 3, 4 or 5 minutes. However, incubators, water baths, etc. may also be used. When aliquots of the sample are taken before the aliquots are heated, they can be placed in different positions in a PCR heating block, with the different desired heating temperatures used in the different positions. The heating step comprises holding the sample (or aliquot thereof) at the desired temperature for a desired period, which may preferably be 20 seconds to 5 minutes, preferably 1, 2, or 3 minutes. The precise holding time depends on the volume of the sample (or aliquot) being heated, and it is within the competencies of the person of ordinary skill in the art to determine an appropriate holding time for their purposes. A preferred holding time is about 3 minutes. When the heating step comprises heating the sample (e.g. the test sample) to a series of temperatures in a temperature range, the result is to obtain a plurality of heated sample aliquots. Thus, preferably the heating step comprises obtaining a plurality of heated sample aliquots by i) heating the non-purified test sample to a series of temperatures in a temperature range and obtaining an aliquot of the sample after heating at each temperature, or ii) obtaining a plurality of non-purified test sample aliquots and heating each sample aliquot to a different temperature within a series of temperatures in a temperatures range to obtain (i.e. produce) a plurality of heated test sample aliquots. In this way, the plurality of heated sample aliquots is the product of the heating step. The number of individual samples used in the heating step may depend on the number of different temperatures which it is desired to expose the sample to. In this instance, each sample aliquot is heated, or has been heated, to only one particular temperature. Each heated sample aliquot may then be processed separately as set out in the subsequent steps of the method. This permits determination of the level of the soluble and / or insoluble protein at each temperature point of the series of temperatures used. Apoptosis is then detected by a difference between the soluble and / or insoluble level of the biomarker protein as compared to a control level in one or more of the heated sample aliquots (or in a protein fraction thereof), i.e. at one or more temperatures (i.e. temperature points) in the series of temperatures used. Thus, the subsequent steps of the method are performed on each of said heated test sample aliquots obtained in the heating step, and wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in at least one of said heated test sample aliquots as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in at least one of said heated test sample aliquots as compared to a control level. Thus, preferably, the methods of the invention comprise a heating step which comprises i) heating the non-purified test sample to a series of temperatures in a temperature range and obtaining an aliquot of the sample after heating at each temperature, or ii) obtaining a plurality of non-purified test sample aliquots and heating each sample aliquot to a different temperature within a series of temperatures in a temperature range to obtain a plurality of heated test sample aliquots; wherein the subsequent steps of the method are performed on each of said heated test sample aliquots, and wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in at least one of said heated test sample aliquots as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in at least one of said heated test sample aliquots as compared to a control level. An “aliquot” of the sample may be of any volume, e.g. 1 or 2 pl. If the sample in question is a cell colony, rather than a liquid sample, then a portion of the colony may be lifted off after each incubation e.g. by placing filter paper on the top of the colony. Thus the invention may comprise heating the sample to a series a range of temperatures and processing and analysing an aliquot taken after incubation at each temperature in order to produce a precipitation curve for one or more biomarker proteins. Optionally, the plurality of heated sample aliquots obtained may be pooled to form a pool of heated sample aliquots, and said pool of heated sample aliquots is thereby the product of the heating step that is processed as set out in the subsequent steps of the method. Throughout this application, any reference to a step being performed on, with or to a heated sample is also to be considered as a reference to said step being performed on, with or to an aliquot thereof, i.e. on, with or to a heated sample aliquot, or a pool thereof. The methods of the invention may comprise a step of lysis before or after the heating step. However, if present then the lysis step is preferably performed after the heating step has been performed. Thus, preferably the sample heated in the heating step contains intact cells. A lysis step is a cell lysis step, i.e. a step of lysing cells in the sample. Such steps are well-known in the art. If a lysis step is performed prior to the heating step, then it is important that the lysis step is non-denaturing, allowing proteins in the sample to retain a native i.e. correctly folded or native-like conformation. This is referred to herein as native lysis. This can be carried out chemically or otherwise using reagents which are well known in the art e.g. mild buffers such as PBS or Tris at physiological pH. The degree of lysis must be sufficient to allow proteins of the cell to pass freely out of the cell. Typically, when dealing with membrane bound proteins, lysis is performed in the presence of detergents or amphiphiles, for example Triton X-100 or dodecylmaltoside, to release the protein from the membrane. The lysis step can alternatively be carried out by freeze thawing the cells or colonies, e.g. in liquid nitrogen. More preferably, lysis is carried out using both native lysis buffer and freeze thawing the cells. The lysis buffer may contain lysozyme, for example at 50-750 pg / ml, or at 100-200 pg / ml. DNAse can also be found in native lysis buffer preferably at 250-750 pg / ml. Native lysis buffer may contain for example 20mM Tris, pH 8, lOOmM NaCl, lysozyme (200pg / ml) and DNAse I (750 ug / ml). Preferably, lysis is performed in the presence of one or more protease inhibitors, which are well-known and widely available. Typically, the cells will be exposed to the lysis buffer for 15-60 minutes, preferably around 30 minutes. The step of freeze thawing is preferably repeated, i.e. two or more cycles, preferably 3 or more cycles of freeze thawing are performed. Preferably, lysis is achieved by a 30 minute incubation at room temperature with lysis buffer and three xlO minutes freeze thawing. A step of lysis may be performed after the heating step. The step is preferably performed before separation of the heated sample (or aliquot) into soluble and insoluble fractions. Again important that the lysis step is non-denaturing (i.e. native lysis), allowing proteins in the sample to retain a native i.e. correctly folded or native-like conformation. If present, a lysis step preferably occurs before the step of determining the soluble / insoluble level of the protein. If present, a lysis step preferably occurs prior to any step of separating soluble from insoluble protein in the heated test sample to produce soluble and insoluble protein fractions. If the biomarker protein is within a cell, and if a lysis step is not performed prior to the heating step, then a lysis step is preferable after the heating step. However, lysis is optional; the detection of soluble / insoluble proteins may be performed within intact cells, e.g. when a cell is fixed and protein-specific imaging probes recognising only the soluble protein region, or alternatively only precipitated protein region, are applied thereto. If fixing cells, it is important that the fixing method used preserves proteins in their native stage. One suitable fixation method comprises the application of formaldehyde (e.g. 4% formaldehyde). If the biomarker protein is inside the cell, then the cell needs to be permeabilized for the probe to access the protein. Permeabilization can be achieved using detergents e.g. NP-40 or triton-x 100. The protein-specific probe may be an antibody conjugated to a fluorophore as described elsewhere herein. Quantification of the level of the probed protein can be performed, e.g. by imaging in a fluorescence microscope or by flow cytometry / FACS. When the present method is carried out on cell colonies, lysis may be carried out directly on those colonies i.e. there is no need to pick the colonies and grow them in liquid culture (although this can be done). In this instance, it is preferred that the separation step is one of filtration. Further, where the method is performed on cell colonies, preferably, the filter paper is overlayed on the colonies to lift the colonies from the semi-solid or solid growth media. Alternatively, filters could be placed on the growth media and cells seeded directly onto the filter, the filter could then simply be lifted off with the colonies already on it. Preferably, the lifting of the colonies in this way can be carried out prior to the lysis step. As indicated above, the lysis can be carried out directly on the colonies on a filter. The filter with colonies attached can be treated with lysis buffer or overlaid on other membranes / filters treated with lysis buffer Typically, the percentage of cells lysed within a sample (e.g. a cell colony or cell culture) during the lysis step is 5-100%. Thus, it is not necessary when performing a step of lysis for all cells within a sample to be lysed. After heating the non-purified test sample, the methods of the invention require determining the level the soluble and / or insoluble biomarker protein in the heated sample (or in one or more aliquots thereof). Preferably, the methods comprise a step of separating soluble from insoluble protein in the product of the heating step to produce soluble and insoluble protein fractions of the test sample. This is discussed in more detail below. However, the separation of soluble and insoluble protein in the heated sample is not essential. For instance, in methods comprising the determination of the soluble level of the biomarker protein, separation of soluble and insoluble protein is not essential. Rather, the level of the soluble protein can be determined in the heated sample itself without separation of the sample into protein fractions. For instance, many antibodies and affinity reagents that recognise the correctly folded structure of the protein will bind the protein with much higher affinity than the unfolded and precipitated protein. Antibodies and other affinity reagents can be linked to magnetic beads or column resin which is mixed with the heat treated non-purified sample. This mix can in a subsequent step be put in an appropriate valve and washed to remove insoluble protein when this does not have high affinity to the affinity reagent. The amount of protein bound to the affinity reagent, can subsequently be measured using for example Bradford techniques, gel electrophoresis, Elisa or surface plasmon resonance detection. It is also possible to use homogenous immune assays where the protein is detected by at least two different antibodies such as alpha-screens, Elisa assays or proximity ligation assays (PLA, Blokzijl, J Intern Med 2010 268;232). When the right antibody combination has been established which specifically recognises the soluble protein, this may give a direct signal for the amount of soluble protein present in the sample, while precipitated protein does not provide a signal. When such separation of the signal from the soluble protein and the insoluble (precipitated) protein can be directly accomplished in a homogenous assay, no physical separation of the soluble and insoluble (precipitated) protein is required to determine the level of the soluble protein. Thus, when the method of detection involves the use of one, or two different, affinity reagents specifically recognise the soluble protein, a heated sample separation step may be excluded. Thus, the invention provides a method for detecting apoptosis, wherein the method comprises the steps of: heating a non-purified test sample comprising a population of cells; and subsequently determining the level of a soluble biomarker protein as defined anywhere else herein in said heated sample by using at least two affinity reagents specific for said soluble biomarker protein, wherein apoptosis is detected by a difference between the level of the soluble biomarker protein as compared to a control level. The affinity reagents are preferably antibodies. However, it will be appreciated that in this aspect of the invention other affinity reagents may be capable of detecting soluble protein in a sample in a similar way to the two antibodies described above. In this aspect, a separation step is unnecessary as the method comprises determination of the level of soluble (or native) protein using two or more affinity reagents (preferably antibodies) that are specific for the soluble form of the protein. By specific is meant that the affinity reagents bind with a higher affinity to the soluble or native form of the protein than to the unfolded or insoluble (precipitated) form of the protein. In this way, such affinity reagents can determine whether soluble or native forms of the biomarker protein are present in a sample after heating. The two or more affinity reagents must together be capable of distinguishing between soluble or native, and unfolded and / or insoluble forms of a biomarker protein and thus must be capable of detecting soluble or native biomarker protein against a background of other proteins (both soluble and insoluble) and any insoluble biomarker protein that may also be present. This form of the method of the invention is particularly advantageous as there is no specific requirement to carry out a separation step, and particularly no separation step may be carried out. In this respect, the methods involve minimal processing steps, which potentially allows automation of the method and an increased throughput, where large numbers of samples can be handled. The “affinity reagent” as used herein thus refers to any reagent which is capable of binding with a higher affinity to a soluble or native form of a biomarker protein than to an unfolded and / or insoluble form of the same protein. An affinity reagent which binds with a higher affinity to a soluble or native form of a protein compared to the unfolded and / or insoluble form of the protein will have a smaller Kd value for its association with the soluble or native target protein than for its association with the unfolded and / or insoluble form of the protein. Particularly, an affinity reagent of the present application may have a Kd value which is at least 100 times smaller with respect to binding to the soluble or native form of the biomarker protein than the Kd with respect to binding to the unfolded and / or insoluble form of the protein. Methods for measuring the Kd values of affinity reagents are well known in the art. Thus, the use of two or more affinity reagents (e.g. antibodies) for the detection of the soluble or native biomarker protein allows the use of affinity reagents which may have a lower specificity for the soluble or native protein than if a single affinity reagent (e.g. antibody) was used alone to detect the soluble or native protein. As two or more affinity reagents (e.g. antibodies) are required to be bound to the soluble or native protein to result in its detection, less specific affinity reagents can be used and can still result in a specific method of detecting the soluble or native protein. Thus, some binding of at least one of the affinity reagents (e.g. antibodies) may occur to the unfolded and / or insoluble form of the biomarker protein, although binding to the soluble or native form of the protein is preferential and association may be at least 100 times greater than association to the unfolded and / or insoluble form of the protein. Similarly, each affinity reagent particularly binds with higher affinity to the soluble or native form of the biomarker protein than to any other protein present in the sample. Preferably, at least one, preferably at least two, of the affinity reagents (e.g. antibodies) used in the method is capable of binding specifically to the soluble or native form of the biomarker protein but not to the unfolded and / or insoluble form of the biomarker protein (or to any other protein). Thus, the reagent(s) may bind specifically to the soluble or native biomarker protein and any binding to the unfolded and / or insoluble biomarker protein may be non-specific and minimal. In embodiments at least one affinity reagent specific for the biomarker protein is capable of binding only to the soluble or native protein and not to the unfolded or insoluble protein, while one or more other affinity reagents specific for the protein may be capable of binding to both forms, but binding to the soluble or native form of the biomarker protein is with higher affinity than binding to the unfolded and / or insoluble form of the protein. Further, the method provides for the use of two or more affinity reagents which bind specifically to the soluble or native form of the biomarker protein but not to the unfolded and / or insoluble form of the protein. In order to distinguish soluble (or native) from unfolded and / or insoluble biomarker protein, the affinity reagent may recognize (particularly specifically recognize) an epitope or sequence of the biomarker protein which is exposed in the soluble or native form of the biomarker protein but not in the unfolded and / or insoluble form of the protein. The two or more affinity reagents recognise different epitopes or sequences on the biomarker protein and thus provide a more specific method of distinguishing soluble or native from unfolded and / or insoluble biomarker protein, than when identifying a single epitope or sequence. (Although with proteins which are homodimers, the affinity reagents may be directed to the same epitope). Thus, both (or more) affinity reagents must be bound to the biomarker protein in order to determine that a soluble or native form of the protein is present. A positive detection of soluble or native biomarker protein is only achieved if both (or more) affinity reagents (e.g. antibodies) are bound. The affinity reagent may be an antibody, antibody fragment, affibody, peptide, aptamer, DARTs or other small molecule that binds to the soluble or native form of a biomarker protein with a higher affinity than to the unfolded and / or insoluble form of the protein. Particularly, at least one affinity reagent is an antibody and more particularly two antibodies are used per biomarker protein to be detected in the method. The invention however also encompasses the use of different affinity reagents e.g. the use of an antibody and another affinity reagent. Since more than one biomarker protein may be analysed in each sample, multiple affinity reagents may be used in each analysis step, where at least 2 affinity reagents are used for each biomarker protein to be detected. Where multiple biomarker proteins are to be analysed, the signals generated by the binding of the two or more affinity reagents to each biomarker protein should be different. The detection of the two or more affinity reagents (particularly two antibodies) may be using a reporter assay which results in a signal change when the two or more affinity reagents are bound to soluble or native biomarker protein. Generally, the two or more affinity reagents are labeled (particularly with different labels) and the close proximity of those labels to one another when bound via the affinity reagents to the soluble or native form of the biomarker protein results in a change in signal e.g. the emission of fluorescence or the production of light or fluorescence at a different wavelength (to the labels when used alone) or the quenching of fluorescence. Such reporter assays are often referred to as proximity reporter assays, e.g. a FRET (fluorescence resonance energy transfer) based method (or a variant thereof, such as BRET (Bioluminescence Resonance Energy Transfer)) may be used in the present method for detection, where the close association of one label (a donor molecule) attached to an affinity reagent, to a second label (an acceptor molecule) attached to an affinity reagent, results in the production or alteration of a signal. Thus, the presence of the two or more labels bound to the soluble or native biomarker protein via the affinity reagents results in a detectable signal change. In embodiments e.g. FRET, transfer of energy from the donor to the acceptor molecule labels may result in the emission of fluorescence by the acceptor molecule. In this way, the labeling of two (or more) affinity reagents (antibodies) which bind to the soluble form of a biomarker protein, with labels which have a signal change when in close proximity to each other, can result in the emission of fluorescence when both are bound to the soluble biomarker protein, thus enabling detection of the soluble target biomarker protein. Donor and acceptor molecules used in FRET / BRET based methods are well known in the art and include pairs such as cyan fluorescent protein and yellow fluorescent protein (both variants of green fluorescent protein); and bioluminescent luciferase and YFP. Such a method requires the binding of both (or more) affinity reagents (e.g. antibodies) to the soluble biomarker protein before detection of any signal (e.g. luminescence) is achieved. In this aspect, one affinity reagent may be labeled with donor molecule and the second affinity reagent may be labeled with the acceptor molecule. The labels which demonstrate a signal change when in close proximity (e.g. when bound via affinity reagents to the soluble biomarker protein), e.g. donor and acceptor molecules, may be coated or comprised within separate bead populations which may then be used to bind to each affinity reagent (antibody). Thus, the beads coated with one of the labels (e.g. the donor molecule) may be used attached to the first affinity reagent e.g. antibody and the beads coated with the second label (e.g. the acceptor molecule) may be used to detect the second affinity reagent (e.g. antibody). Each bead population (e.g. donor or acceptor) may be conjugated with a further reagent to allow binding to either the first or second affinity reagent (e.g. antibodies). For example, a bead population (donor or acceptor) may be streptavidin coated to allow binding to biotinylated affinity reagent (e.g. antibody) or maybe conjugated to protein A to allow binding to an antibody affinity reagent. Methods of attaching beads to affinity reagents such as antibodies are well known in the art. It will be appreciated that the two or more affinity reagents (e.g. antibodies) may be labeled (e.g. with donor or acceptor molecules) prior to addition to the sample or after addition to the sample. However, particularly, the affinity reagents (antibodies) may be labeled before addition to the sample. The AlphaScreen Surefire assay format (Perkin Elmer) may be particularly used in the method of the invention, where other antibodies may be attached to the beads provided (see Osmond et al, Analytical biochemistry, 403, 94-9101, 2010, incorporated herein by reference). Other methods for detection of the bound affinity reagents include proximity ligation assays (such as Duolink from Olink) and ELISAs. Methods for producing affinity reagents, such as antibodies which may bind to the soluble form of a protein but not the insoluble form of a protein are known in the art. For example, the study of the 3D structure of a soluble and insoluble protein can enable determination of epitopes which are exposed on the soluble form but not the insoluble form. Antibodies or peptides which bind to such epitopes can then be produced using standard methods. Further, methods are known which can be used to identify antibody pairs that bind to a protein e.g. methods employing surface plasmon resonance biosensors or ELISA. Further, Bembenek et al (Analytical Biochemistry, 408, 2011, 321-327, incorporated herein by reference), reported a bead-based screening method using antibody capture on Protein A Alphascreen beads to analyse and select pairs of antibodies capable of binding to the same target. Thus, according to the methods of the invention it is possible to determine the level of the soluble biomarker protein in the whole heated sample (the product of the heating step), i.e. a step of separating soluble and insoluble protein fractions is not necessary. However, the methods of the invention preferably comprise a step of separating soluble from insoluble protein in the product of the heating step (i.e. in the heated sample or aliquots thereof), to produce soluble and insoluble protein fractions of said sample, and then determining the levels of the biomarker protein in either or both of the soluble and insoluble protein fractions. This separation step achieves the physical separation of the proteins that remain stable (i.e. are in the soluble fraction) from the proteins that are destabilized (i.e. are in the insoluble fraction) after the heating step. Stable as used herein means “thermally stable”, i.e. not denatured or “destabilized” due to the heating of the non-purified sample in the earlier step. Thus, preferably, the present invention provides a method for detecting apoptosis, wherein the method comprises the steps of: a) heating a non-purified test sample comprising said population of cells; b) separating soluble from insoluble protein in the product of step a) to produce soluble and insoluble protein fractions of said heated test sample; and c) determining the level of the biomarker protein in either or both of said soluble and insoluble protein fractions of said heated test sample, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level. Alternatively viewed, apoptosis is detected by an increase in the level of the biomarker protein in the soluble protein fraction as compared to a control level, and / or by a decrease in the level of the biomarker protein in the insoluble protein fraction as compared to a control level. The separation step b) is a simple step of separating soluble from insoluble proteins to produce (i.e. obtain) soluble and insoluble proteins fractions after heating. Soluble proteins are associated with being thermally stable at the temperature applied to the sample. The separation step is to discriminate between soluble and insoluble proteins. The separation step can involve any separation method which is capable of separating soluble from insoluble protein. For example, a step of centrifugation, chromatography or filtration may be used. Preferably the separation step is centrifugation or filtration. Any suitable centrifugation step for separating soluble and insoluble proteins in a heated sample may be used, and such steps are within the general competencies of the person of ordinary skill in the art. Centrifugation can typically be carried out between 100g and 50000g. Centrifugation is preferably carried out above 2000g, e.g. between 2000g and 50000g, or between 2000g and 20000g. The duration of centrifugation can be from 1 minute (typically at least 10 minutes) to at least 1 hour, where the duration required generally decreases as the centrifugal force increases. Particularly suitable conditions for providing only soluble proteins in the resultant supernatant include 30 minutes at 3000g or 20 minutes at 20000g. A filter can be used to separate soluble from insoluble proteins where soluble proteins will pass through a filter. Standard filter membranes can be used for filtering heated samples where the filters will typically have a pore size from 0.015 pm to 12 pm, preferably from 0.35 to 1.2 pm, more preferably from 0.45 pm to 0.8 pm. Preferably the filters have pore sizes below 4.0 pm, typically below 2.0 pm, more preferably below 1.0 pm. Suitable filters are Super and GH polypro (from Pall) and Nucleopore (From Whatman). It will be appreciated that proteins of interest from eukaryotic and prokaryotic samples and from different cell types may require the use of filters with different pore sizes. Selection of a suitable filter is well within the competency of someone skilled in this field. For example, it is possible to select an appropriate pore size, by using a set of test proteins for the desired cell type or sample and investigating their behaviour with filters of varying pore sizes. Proteins can simply be allowed to pass through the filter, possibly as a result of capillary action. Alternatively, force may be applied vertically on the filter paper, wherein such forces can include the application of pressure or vacuum. In all aspects and embodiments, the methods of the invention preferably comprise determining and comparing the levels of soluble protein or groups of proteins. In the aspects and embodiments of the invention that comprise a separation step, the separation step results in separate soluble and insoluble protein fractions of the heated sample. According to the methods of the invention it is possible to analyse either (or both) the insoluble or soluble protein fractions in the subsequent method steps. Thus the level of the biomarker protein(s) may be determined in either or both of the soluble and insoluble protein fractions. Preferably, the methods of the invention involve a step (c) of determining the level of the biomarker protein(s) in the soluble protein fraction(s) of the test sample. In other words, preferably the soluble protein fraction of step b) is analysed. Hence, if step b) comprises a centrifugation separation step, then the supernatant can be analysed to determine the level of biomarker protein(s). If step b) comprises a filtration separation step, then the filtrate can be analysed to determine the level of the biomarker protein(s). If the insoluble fraction is analysed, then it is preferably solubilised prior to analysis. Steps for solubilizing an insoluble protein fraction are well-known in the field, and any suitable solubilization step may be performed. For example, the insoluble proteins in the insoluble protein fraction may be solubilized with detergents or combinations of detergents, including sodium dodecyl sulfate, thiol reducing agents (such as 2-mercaptoethanol and dithiothreitol) and chaotropic agents such as urea, etc. The protein fraction may be subsequently dissolved in loading buffer prior to application to separation gels. It is important to note that if an insoluble fraction is solubilized for analysis, the level of insoluble protein in the heated test sample is still being determined due to the previous separation of the insoluble and soluble protein fractions. The methods of the invention comprise determining the level of soluble or insoluble biomarker protein in said heated sample (or sample aliquot), or protein fraction thereof. By “determining the level” of a biomarker protein means determining the abundance of the protein in the sample, aliquot or fraction concerned. Determining the level of a protein may comprise quantification of the protein, i.e. quantitative determination of the amount or concentration of the protein. The determination of the level of a biomarker protein is the relative level of the biomarker protein as compared to a control level, which is the level of the same protein in a control sample. As explained below, mass spectrometry is a preferred method for determining the level of soluble or insoluble biomarker protein. It is within the general competencies of the person of ordinary skill in the art to interpret peaks in mass spectrometry data to determine the level of particular proteins therein. Many well-known and widely available computational tools exist for this purpose, and any suitable tool may be used. Determining the level of the biomarker protein(s) may be performed by any suitable technique. Preferably, the technique does not involve the labelling or tagging of the biomarker protein prior to the performance of the heating step (e.g. step a). However, tagging or labelling of the protein in the cell before the heating step may be performed, and may be desirable in some instances, e.g. to assist with quantification of the protein after heating (and optionally lysis) e.g. with an affinity reagent. Tagging and labelling of the protein after heating step (step a)) has been performed is also possible. Preferably, if determining the level of the biomarker protein in a soluble or insoluble protein fraction, then the level of the biomarker protein is determined by mass spectrometry or by affinity binding. Determination of the level of a biomarker protein by affinity binding refers to the level of a biomarker protein being determined by affinity binding between the biomarker protein and an affinity reagent, for example an antibody, antibody fragment or affibody (non-Ab based protein binding partner) known to recognise the biomarker protein. Preferably, antibodies, monoclonal or polyclonal, against the biomarker protein may be used. Antibody-based methods allow for rapid and reliable analysis of a wide variety of peptides and proteins, including those which themselves possess no catalytic activity. Such methods are routine in the field. It is within the competencies of the person of ordinary skill in the art to generate, select and / or obtain one or more affinity reagents against a desired biomarker protein, and to use the same in order to detect the level thereof in a sample or protein fraction. Affinity reagents are discussed above in the context of two such agents being used to detect the soluble biomarker protein in a heated sample without performance of a separation step, and the same preferred and optional features of such affinity reagents and their use apply mutatis mutandis to the affinity reagents and their use when applied to a soluble or insoluble protein fraction. If the soluble and insoluble protein fractions have been separated, the affinity reagent used to detect the biomarker protein in either of said fractions need not be specific for the soluble or insoluble form of the protein, it is sufficient for it to be specific for the biomarker protein per se. In general, methods exploiting affinity binding to detect specific proteins rely on the coupling of the affinity regent to an enzyme, fluorophore or DNA to allow quantification of the amount of the protein of interest. Alternatively, the detection reagent can be coupled to a secondary affinity reagent that binds to the protein-specific affinity reagent. Plate based assay formats where the affinity reagent is coupled to an enzyme (enzyme-linked immunosorbent assay (ELISA)) can be used. Homogenous assays, such as AlphaLISA and PEA, in which two affinity reagents binding the same protein region are required, are useful since this increases the specificity and eliminates the need of washing steps. Imaging formats and FACS based techniques have the advantage of allowing analysis of protein levels in individual cells. Determination of the level of a biomarker protein by affinity binding may of course be used if the identity (amino acid sequence) of the biomarker protein is known. However, even if the identity of the biomarker protein is unknown, affinity based methods may still be used. For instance, antibody arrays may be used to determine the level of multiple proteins in a sample simultaneously; such methods allow for rapid and reliable analysis. Protein affinity arrays, based on specific antibodies, can be used to quantify proteins. Preferably, the level of the biomarker protein is determined by mass spectrometry. Mass spectrometry is particularly preferred when the levels of many proteins are to be determined, e.g. in methods in which levels of a plurality of biomarker proteins, or groups of biomarker proteins, are to be determined. Mass spectrometry analysis typically requires a separation step as described above to be performed and the subsequent analysis of the soluble fraction, or of the insoluble fraction, as described above. Mass spectrometry analysis of an insoluble protein fraction requires a step of solubilization of the fraction prior to analysis, as described above. Preferably, step c) of the method preferably comprises subjecting either or both the soluble and insoluble protein fractions of step b) to mass spectrometry analysis to produce mass spectrometry data, and analysing the mass spectrometry data to determine the level of the biomarker protein(s) in said protein fraction(s) of said heated test sample. Methods of mass spectrometry, and analysis of mass spectrometry data to determine levels of protein sequences therein, are well known in the field, and any suitable method may be used. Preferred methods are described in the Examples. Semi-quantitative or preferably quantitative mass spectrometry (MS) may be used. A preferred method is liquid chromatography MS (LC-MS). Mass spectrometry methods can be used to measure a large fraction of the proteome, but can also be targeted to detect only a smaller number of proteins, if desired. Steps of sample preparation for mass spectrometry analysis are well-known in the field and any such steps may be used. It is within the competencies of the person of ordinary skill in the art to prepare samples for mass spectrometry analysis. Preferred methods are described in the Examples. The methods thus may comprise a step of preparing the protein fraction(s) of step b) for mass spectrometry analysis. When performing mass spectrometry on protein-containing samples it is typical to denature and subsequently digest proteins in the sample (or protein fraction thereof) into smaller fragments (i.e. peptides). This is typically done via proteolysis, e.g. trypsinisation, i.e. trypsin digestion) of the proteins in the sample. For instance, trypsinization may be performed at 37°C overnight. Preferred methods are as described in the Examples. After digestion, the resulting peptides may be labelled to facilitate detection by mass spectrometry. Again, MS-labelling strategies are well known in the art, and any suitable method may be used. Preferred methods of labelling include ITRAQ, or Isobaric Tandem Mass Tags (TMT), e.g. TNT-lOplex or (TMTpro)-16plex. The TMT-lOplex strategy is preferred as it permits 10 temperatures to be measured and normalized for each sample in the same experiment. Prior to MS measurement, samples (e.g. comprising labelled trypsinisation product peptides) may be pre-fractionated, e.g. via liquid chromatography. Suitable techniques are well known in the field. The MS measurement would be performed using a mass spectrometry instrument, such as an Orbitrap Q Executive or Fusion (Thermo-Fisher Scientific). Full scan MS spectra may be acquired, e.g. in the range of 375-1500 m / z. Mass spectrometry analysis results in mass spectrometry data, which can be analysed by well-known and commonly used data processing programs to indicate the levels of each of the peptide fragments, and so the levels of the biomarker proteins, present in the analysed sample. For instance, in a Fourier transform ion cyclotron resonance experiment using an orbitrap instrument, typically 1000-10 000 proteins can be detected simultaneously in a complex sample (e.g. a cell lysate). As explained above apoptosis is detected by an observed i) increase in the level of at least one soluble biomarker protein as defined anywhere herein in the heated test sample or aliquot(s), or soluble protein fraction thereof, as compared to a control level, or i) decrease in the level of at least one insoluble biomarker protein as defined anywhere herein in an insoluble protein fraction of the heated test sample or heated test sample aliquot(s). Measurement of the soluble biomarker protein in the soluble protein fraction of the heated test sample or heated test sample aliquot(s) is preferred. Apoptosis is determined by a difference between the soluble / insoluble level of the biomarker in the test sample as compared to a control level (i.e. in the level in a control sample) because this indicates that the conditions of the test sample have caused a change (shift) in the thermal stability of the biomarker protein, which is indicative of apoptosis. The present invention is predicated on the realisation that apoptosis causes a thermal stability change in the biomarker proteins defined herein. In an apoptotic cell, the present biomarker proteins are thermally stable to a greater extent than they are thermally stable in a non-apoptotic cell. Without wishing to be bound by theory, the present inventors consider that the induction and progress of apoptosis alters a variety of intra- and inter-molecular interactions involving the protein biomarkers, including protein cleavage and altered association / disassociation of proteins with protein and non-protein binding partners, largely driven by apoptotic cell signalling events. As a result, the biomarker proteins may have greater thermal stability at a given temperature as compared to their thermal stability in a non-apoptotic cell. It was not previously known or expected that the CETSA methodology was sufficiently sensitive for use in analysing thermal stability shifts in proteins due to cellular apoptosis, but the present inventors have demonstrated this. The present inventors have identified a panel of protein biomarkers of apoptosis, the thermal stability of which is altered “shifted” in apoptotic cells as compared to in non-apoptotic cells. Thus, in a CETSA assay, a change in the soluble or insoluble level of one or more of the biomarker proteins as compared to a control level is indicative of apoptosis. The level of a biomarker protein as used herein is a relative level, i.e. the difference in the level of the biomarker protein in the test sample as compared to in the control sample. If, at a given temperature point, the amount of soluble / insoluble biomarker protein in the test sample is different (i.e. has “shifted”) as compared to in the control sample, then the conditions of the test sample have caused a “stability shift” for that protein. The present inventors have determined that when the present biomarker proteins are shifted, this is due to apoptosis of cells in the test sample, i.e. a greater degree of apoptosis in the test sample as compared to the control. In the present methods, apoptosis is detected when a stability shift is observed for one or more of the biomarker proteins defined anywhere herein, specifically when an increase in stability is observed. If a difference between the level of a biomarker protein in the test sample as compared to in the control sample is observed, then typically it is observed at temperature points above the initial melting point of the biomarker protein, but not necessarily at lower temperatures. The difference between the determined level of the biomarker protein (i.e. in the test sample or aliquot, or protein fraction thereof) and the control level is a measurable difference, preferably a significant difference, preferably a statistically significant difference. Thus, the increase in the level of the soluble biomarker protein(s) is preferably a statistically significant increase. Appropriate methods of determining statistical significance are well known and documented in the art and any of these may be used. Such methods may be based on calculation of P values or alternatively standard deviation or related methods using replica measurements. In any statistical analysis referred to herein, preferably the statistically significant difference over the control level has a probability (P-value) value of <0.1, preferably <0.05. The soluble level of the biomarker protein may be at least 25%, preferably at least 50%, more preferably at least 75% or at least 100% greater than the control level in order that apoptosis be detected. The determined soluble level may be many fold greater than the control level, for instance at least 2 fold, at least 3 fold, at least 4 fold or at least 5 fold, or greater. The insoluble level of the biomarker protein may be at least 25%, preferably at least 50%, more preferably at least 75%, at least 100% lower, than the control level in order that apoptosis be detected. The determined insoluble level may be many fold lower than the control level, for instance at least 2 fold, at least 3 fold, at least 4 fold or at least 5 fold, or more. The relative level of a biomarker protein (i.e. its “shift”) can also be expressed in the form of a ratio or “fold change”. These provide numerical values for the difference between the level of the biomarker protein (soluble or insoluble) in the heated test sample (or aliquot or protein fraction thereof) as compared to the level of the same biomarker protein (likewise soluble or insoluble) in the heated control sample (or aliquot or protein fraction thereof). Thus, a ratio or “fold change”, indicates the extent of the difference between the test and control levels for a given biomarker protein. The ratio may be expressed as the ratio of the test level of the biomarker protein to the control level of the biomarker protein, or vice versa. The fold change (or ratio) may be expressed as the test level of the biomarker protein divided by the control level of the biomarker protein, or vice versa. The fold change may be expressed on a log scale (“log fold change” or “log FC”, expressed using any desired base unit. Preferably, the fold change is expressed as a "log2 fold change", which is calculated according to the formula log2(FC) = log2 = log2(a) — log2(b), wherein a and b are the test and control levels of the biomarker protein concerned, respectively As explained above, the test and control levels may be determined computationally, e.g. from mass spectrometry data. Software for determining and displaying the ratio and / or fold changes discussed above is also widely available. Measurement of individual protein abundance may be computed from raw mass spectrometry data using proteomics software, which is well-known and widely available, such as Proteome Discoverer or MaxQuant. The data obtained may be normalized, e.g. by log2 transformation. The log2 fold changes can then be obtained by computing the difference between test and control samples. If the soluble level of a biomarker protein in a heated test sample is raised, then it follows that the insoluble level of the biomarker protein in the heated test sample will be reduced, as compared to the control, and vice versa. Thus, when the method comprises determining the level of both soluble and insoluble biomarker protein (e.g. that in the soluble protein fraction and in the insoluble protein fraction of a heated test sample or aliquot), then the level of the biomarker protein can be expressed as the ratio of the determined soluble level to the determined insoluble level, or vice versa. This test level ratio of the biomarker protein can be compared to a control level ratio (which is determined in the same way using the control sample), and apoptosis is detected by a difference between the two ratios. It is important to note that the total amount of the biomarker protein in the sample does not change significantly upon apoptosis. CETSA assays are based on thermal stability changes in a protein, wherein heating results in altered levels of the soluble and insoluble forms of the protein being detected (e.g. in the soluble and insoluble protein fractions of the sample) as compared to in heated control samples. But the amount of the protein is the sample is not altered. The observed increases in the level of the protein are not due to increased expression of the protein; this can be confirmed / controlled via standard quantitative proteomics, which is within the competencies of the person of ordinary skill in the art. As explained above, the determination of the level of a biomarker protein comprises determining the relative level of the biomarker protein as compared to a control level, which by definition is a control level of the same biomarker protein. This is the case for all biomarker proteins assessed, and for insoluble and soluble levels alike. A control level is the level of the biomarker protein determined in a corresponding heated control sample (or aliquot) or protein fraction thereof. The control sample is a non-purified control sample. The control level is determined in a heated non-purified control sample (or aliquot) or protein fraction thereof, using the same (corresponding) method steps as used to determine the level of the biomarker protein in the test sample. Thus, the determined level of a biomarker protein in the heated test sample or test sample aliquot, or protein fraction thereof, is the relative level of said biomarker protein as compared to a control level of said biomarker protein, and apoptosis is determined by a difference between the relative level of the biomarker protein as compared to the control level. The person of ordinary skill in the art understands that comparison to a control level means comparison of the level of same biomarker protein in the same insoluble or soluble form, in the same (i.e. corresponding) sample or protein fraction type of a control sample. By way of example, the level of the soluble biomarker protein in the heated test sample would be compared to the level of the same soluble biomarker protein in the heated control sample, the level of soluble biomarker protein in the soluble protein fraction of the heated test sample would be compared to the level of the same soluble biomarker protein in the soluble protein fraction of the heated control sample, and the level of insoluble biomarker protein in the insoluble protein fraction of the heated test sample would be compared to the level of the same insoluble biomarker protein in the insoluble protein fraction of the heated control sample. Thus, when a soluble fraction of a test sample is analysed, the soluble fraction of the control sample is also analysed. Further, insoluble fractions are compared with each other from the test and control samples. Thus, the level of soluble / insoluble biomarker protein determined is the relative level (relative abundance) of the soluble / insoluble biomarker protein as compared to levels of said biomarker protein in a suitable control (sample). Preferably the control, i.e. the control sample / aliquot / fraction, referred to in relation to one biomarker protein is the same control, i.e. the same control sample / aliquot / fraction, referred to in relation to further biomarker proteins assessed as part of the method. It is within the competencies of the person of ordinary skill in the art to identify a suitable control sample. A control sample is also a non-purified sample. The control sample is preferably of the same or similar type (i.e. corresponding) type as the test sample, unless differences in the sample type are to be analysed for their effects on apoptosis. Thus, for example, if a tissue sample from a tumour is used as the test sample, preferably the same tissue sample type is used as the control sample, e.g. a tissue sample from a tumour. If the two samples are subject or patient samples, then they are preferably obtained from the same subject or patient. They may be obtained from the same subject or patient at different times. Furthermore, both the test and control samples will preferably have been exposed to the same or similar heating and separation steps e.g. heated at the same temperature(s) or a at similar temperature(s), e.g. within 0.5-2°C, and / or treated with the same separation, lysis and other steps, if performed. The test and control sample may comprise the same sample at different time points, or may be subject samples obtained at different time points. The test and control samples may comprise the same cell types, unless it is desired to analyse apoptosis across differing cell types. Thus, the test and control samples are largely similar or identical, and have been similarly or identically treated. However, the test sample differs from the control sample in one respect, or has been treated differently in one respect, such that a comparison can be made between the test and control samples in order to analyse the effects of that difference on apoptosis. The person of ordinary skill in the art understands that a test sample by definition comprises a feature (or component) that is absent from a control sample, or has been exposed to a condition or treatment step to which the control sample has not been exposed, or vice versa. Thus, preferably, the test sample of the invention has been exposed to a test condition, to which the control sample has not been exposed. Thus, the apoptosis detected by the methods of the invention is preferably apoptosis due to, or induced by, said test condition. The control level is preferably obtained from a control sample that is not apoptotic, i.e. comprises a majority of cells that are non-apoptotic cells. In such a control sample, essentially all cells are non-apoptotic, i.e. there may de minimis levels of apoptosis in the cell population in the control sample, but the level is insignificant. Alternatively viewed, preferably the test sample has been exposed to a first (test) condition and the control sample has not been exposed to said first (test) condition. By way of non-limiting example, the test sample and the control sample may have been differentially treated; exposed to different conditions; obtained from different subjects, different cell cultures or different tissue cultures; or obtained from the same subject, cell culture or tissue culture at different time points. The test sample and the control sample may comprise cells of a different genotype or phenotype. Optionally: i) the test sample has been exposed to a test molecule (e.g. a drug, a physiological metabolite or a cytokine) and the control sample has not been exposed to said test molecule or vice versa, ii) the test sample has been obtained from a drug-treated subject and the control sample has been obtained from a subject that has not been drug-treated or vice versa', iii) the test sample and the control sample have been treated with different drugs or are obtained from subjects that have been treated with different drugs; iv) the control sample and the test sample have been obtained from the same subject at different time points; v) the control sample and the test sample have been obtained from the same subject before and after drug treatment, respectively, or vice versa. vi) the test sample is a patient sample and the control sample is a healthy control sample or vice versa', vii) the test sample and control sample are different biological samples or comprise different cell lines; viii) the test sample is a diseased sample and the control sample is a healthy control sample or vice versa', ix) the test sample and control sample are obtained at two different time points from the same cell culture or tissue culture; or x) the test sample comprises cells comprising a genetic modification and the control sample comprise cells lacking said genetic modification, or vice versa. Thus, the test condition may be treatment with a test molecule, and is preferably drug treatment. The "test molecule" as used herein may be a protein, polypeptide, peptide, RNA, or DNA molecule. The test molecule may be a drug or pharmaceutical product, a cell metabolite or a hormone e.g. in serum. The test molecule may be naturally occurring or may be synthetically or recombinantly produced. The test molecule may alternatively be a test composition, e.g. comprising a mixture of molecules as defined herein. Plant extracts are an example of a preferred test composition. Thus, the test condition may be treatment with a test molecule, drug treatment. The "test molecule" as used herein may be a protein, polypeptide, peptide, RNA, or DNA molecule. The test molecule may be a drug or pharmaceutical product, a cell metabolite or a hormone e.g. in serum. The test molecule is preferably a drug, more preferably an anti-cancer drug. The test molecule may be naturally occurring or may be synthetically or recombinantly produced. The test molecule may alternatively be a test composition, e.g. comprising a mixture of molecules as defined herein. Plant extracts are an example of a preferred test composition. The test condition may also be that further (different) cells are added to the test sample, i.e. the cells in the test sample are exposed to cells of a further (different) type, such as immune cells, in order to detect apoptosis induced by said further cells. Preferably, the added further cells are washed off before any step of lysis step is performed. Alternatively, the added further cells may have been grown in labelled amino acids, which results in the proteins and peptides of such further cells having different molecular weights from those in the cells of the test sample. This allows proteins and peptides from the different cell types to be distinguished by mass-spectrometry. Preferably, the method of the invention comprises a step of exposing the test sample to said test condition, said step being performed prior to the heating step (e.g. step a)). Preferably, the test condition is a step of adding (i.e. administering) a test molecule, as defined anywhere herein, to the non-purified test sample, i.e. a step of exposing the cells (cell population) within the non-purified test sample to the test molecule. Preferably this is an in vitro step. Thus, preferably, the non-purified test sample is a sample which has been exposed to, i.e. to which has been added (i.e. administered), a test molecule as defined anywhere herein. The methods may comprise administering said test molecule to a subject or patient prior to obtaining a sample from said subject or patient, wherein the obtained sample is the nonpurified test sample of use in the method. Thus, preferably the non-purified test sample is a sample which has been obtained from a subject or patient to whom a test molecule has been administered. Alternatively viewed, the non-purified sample on which the steps of the method are performed comprises said test molecule. Again, the control level is that determined (previously, or as part of the method) in a control sample that has not been exposed to said test condition, e.g. which has not been exposed to said test molecule (e.g. drug). The step of heating the non-purified test sample, and the subsequent determination of the level of the biomarker protein(s), whether soluble and / or insoluble, may be performed at any time point after exposure of the non-purified test sample to the test condition (e.g. at any time point after drug treatment). In other words, the method of the invention may be performed at any time point after exposure of the non-purified test sample to the test condition (e.g. at any time point after drug treatment). Preferably, the method is performed within 24 hours, more preferably within 18 hours, more preferably within 12 hours, more preferably within 8 hours of exposure of the non-purified test sample to the test condition. Apoptotic processes may be initiated within 1 hours of exposure to the test condition, and apoptosis may advantageously be detected within an 8 hour timeframe, e.g. after 1 hour, 2 hours, 4 hours, 6 hours and / or 8 hours of exposing the non-purified test sample to the test condition. Again, these time frames are preferably after drug administration. Thus, the methods of the invention may also be considered as methods of detecting the initiation of apoptosis. Preferably, when detecting the initiation of apoptosis, the method is performed within the above-discussed periods following exposure of the non-purified test sample to the test condition. Typically, where the test molecule is present extracellularly e.g. in solution, this may be simply added to the non-purified sample. Alternatively, the test molecule may be expressed recombinantly from a vector encoding the test molecule. The step of adding the test molecule may therefore involve transforming or transfecting a cellular sample with the vector encoding the test molecule and / or inducing expression of the test molecule from the vector in a cellular sample once transformation or transfection has been carried out. The step of adding the test molecule further includes inducing expression of a test molecule encoded by a gene naturally occurring in a cellular sample. Further, the method may require an extracellular test molecule to be transported into the cell. Alternatively, or additionally, a step of lysis may be carried out before, simultaneously or after the test molecule has been added. Such a lysis step allows contact between the proteins in the cells in the sample and the test molecule. Thus, any necessary lysis step is generally carried out before the separation step of the method of the invention. The methods of the invention may comprise as step of repeating the earlier steps with one or more (e.g. 2 or more, preferably 3 or 4 or more) different concentrations of the test molecule. As mentioned above, the level of the soluble / insoluble biomarker protein that is determined is the level of the soluble / insoluble biomarker protein relative to a control level of the same soluble / insoluble biomarker protein. The methods of the invention may comprise performing the indicated steps on the test sample, and on a control sample, in order to determine the levels of biomarker protein in each. If the method of the invention comprises repeating the steps of the method on a non-purified control sample, then said repeated steps can be performed simultaneously to, before or after the steps performed with the non-purified test sample. The test and control samples may therefore be obtained and treated at different times, and simultaneous or directly sequential processing is not necessary (although can be done). The period of time between performing the steps on the test and control samples may be any period of time, e.g. a few minutes (e.g. between 1-10 minutes), hours (e.g. from 1, 2, 5, 10 hours), months (from 1, 2, 5, 10 months) or years (from 1, 2, 3, 4 or 5 years). Particularly, it will be appreciated that in the case where the samples are obtained from the same patient, e.g. at different time points, a significant period of time may occur between collection if the samples and / or performance of the method thereon. However, performance of the method steps on a control sample is not essential for the performance of the method. Rather, the level of biomarker protein determined as part of the method of the invention may be compared to a control level that has been obtained previously, e.g. via a “reference database” or “control database”. In other words, the data obtained with the control sample may be in the form of reference data, e.g. a reference database, and the methods of the invention need not comprise the performance of steps on a control sample. Preferably, the method comprises the steps of: a) heating a non-purified test sample comprising a population of cells; b) preferably separating soluble from insoluble protein in the product of step a) to produce soluble and insoluble protein fractions of said heated test sample; and c) determining the level of a soluble biomarker protein, as defined anywhere else herein, in said heated test sample or in a soluble protein fraction thereof, and / or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in said heated test sample or soluble protein fraction thereof as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in the insoluble protein fraction of said heated test sample as compared to a control level. Preferably, the method is for detecting apoptosis, wherein the method comprises the steps of: a) heating a non-purified test sample comprising a population of cells, wherein said heating step comprises i) heating the non-purified test sample to a series of temperatures in a temperature range and obtaining an aliquot of the sample after heating at each temperature, or ii) obtaining a plurality of non-purified test sample aliquots and heating each sample aliquot to a different temperature within a series of temperatures in a temperature range, to obtain a plurality of heated test sample aliquots; b) preferably separating soluble from insoluble protein in each of the heated test sample aliquots to produce soluble and insoluble protein fractions thereof; and c) determining, for each heated test sample aliquot, the level of a soluble biomarker protein, as defined anywhere else herein, in said heated test sample aliquot or in a soluble protein fraction thereof, and / or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample aliquot, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in said heated test sample aliquot or soluble protein fraction thereof as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in the insoluble protein fraction of said heated test sample as compared to a control level. As explained above, the level of a soluble biomarker can be determined in a heated sample (or heated sample aliquot) or in the soluble protein fraction thereof, and the level of insoluble biomarker protein can be determined in the insoluble protein fraction of a heated sample (or heated sample aliquot). As mentioned above, in all aspects and embodiments, preferably step c) comprises subjecting either or both the soluble and insoluble protein fractions of step b), preferably the soluble fraction, to mass spectrometry analysis to produce mass spectrometry data; and analysing the mass spectrometry data to determine said levels of biomarker protein. As explained above, the methods of the invention preferably comprise heating the non-purified sample to a series of temperatures in a temperature range to obtain a plurality of heated test sample aliquots, which are then each processed by the subsequent steps of any method herein. Thus, it follows, that if a plurality of heated test sample aliquots is obtained then the methods preferably comprise determining the relative level of the biomarker protein in each heated test sample aliquot to the level of the same biomarker protein in the equivalent aliquot of a suitable control sample heated to the same temperature point. It is then possible to prepare a plot of the relative levels discussed above (e.g. fold changes) at each temperature point (see e.g. Figures 1, 2 and 5 herein). Apoptosis is detected by a difference between the determined level of the soluble and / or insoluble biomarker protein and a control level using any one or more the heated sample aliquots, i.e. at any one or more temperature points. In methods in which a plurality of heated test sample aliquots is obtained and processed, it is of course the case that comparison to a control level would be performed for each heated sample aliquot by comparison to an equivalent (or corresponding) heated control sample aliquot (i.e. which had been heated to the same temperature point in the series), or protein fraction thereof. The relative levels of a given biomarker protein at two or more, preferably all, of the temperature points used in the method can be summed to provide a single “summed relative level” (e.g. a “summed fold change”). Thus, optionally, the relative levels of a biomarker protein determined using each heated test sample aliquot (i.e. the level at each temperature as compared to a control level at said temperature) are summed to provide a “summed relative level” for that biomarker protein (e.g. a summed fold change). Apoptosis may be determined by a significant summed fold change for one or more of the biomarker proteins as defined anywhere else herein. The present invention also provides the following methods. The preferred and optional features and embodiments described above, i.e. in relation to the methods for detecting apoptosis of the invention, apply mutatis mutandis to each and all of the further methods described below. The methods of the present invention are useful for the identification of compounds that inhibit or stimulate apoptosis. Such compounds may be useful therapeutics, e.g. in neurodegenerative disease and oncology therapies. Thus, in a further aspect, the present invention provides a method for identifying an apoptosis-inducing agent, wherein the method comprises the steps of: heating a non-purified test sample comprising a population of cells that has been exposed to a test compound; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, wherein the test compound is identified as an apoptosis inducing agent if apoptosis is detected, i.e. by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level, and wherein said biomarker protein is as defined elsewhere herein. In this aspect, the control sample has not been exposed to said test compound, or has been exposed to a lower amount (concentration) of said test compound. These methods can be alternatively viewed as methods of assessing the ability of a test compound to induce or enhance apoptosis of cells in a non-purified test sample These methods may comprise a step before the heating step of exposing the non-purified test sample to said test compound. Preferably the compound is a test compound as defined elsewhere herein. Preferably the compound is a drug, e.g. an anticancer agent. Preferably the method is a high-throughput (drug) screen. Alternatively, the method may be for identifying an apoptosis-inhibiting agent, wherein the method comprises heating a first non-purified test sample comprising a population of cells that has been exposed to a test compound and to an apoptosis-inducing agent; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, determining the increase in the level of the soluble biomarker protein as compared to a control level, and / or the decrease in the level of the insoluble biomarker protein as compared to a control level, and comparing the determined increase and / or decrease with the increase and / or decrease determined with a second non-purified test sample that has been exposed to an apoptosisinducing agent but not to the test compound and processed via the same steps, wherein the test compound is identified as an apoptosis inducing agent if the increase in the level of soluble biomarker protein in the first non-purified test sample as compared to a control level is reduced as compared to that determined with the second non-purified test sample as compared to a control level, and / or wherein the test compound is identified as an apoptosis inducing agent if the decrease in the level of insoluble biomarker protein in the first non-purified test sample as compared to a control level is reduced as compared to that determined with the second non-purified test sample as compared to a control level. Alternatively, the method may be for identifying an apoptosis-inhibiting agent, wherein the method comprises heating a first non-purified test sample comprising a population of cells that has been exposed to a test compound and to an apoptosis-inducing agent; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, determining the increase in the level of the soluble biomarker protein as compared to a control level, and / or the decrease in the level of the insoluble biomarker protein as compared to a control level, wherein the test compound is identified as an apoptosis inducing agent by a decrease in the level of soluble biomarker protein as compared to a control level, and / or by an increase in the level of the insoluble biomarker protein as compared to a control level, wherein the control level is determined using a sample comprising apoptotic cells. The biomarker protein may be as defined anywhere else herein. The control sample comprising apoptotic cells may be prepared by exposing a non-purified sample comprising cells to a known apoptosis-inducing agent, many of which are well known in the field, e.g. venetoclax. In other words, in these methods, the test compound is identified as an apoptosisinhibiting agent if apoptosis is not detected, or is detected to a lesser extent than in a control sample comprising apoptotic cells. These methods may comprise a step before the heating step of exposing the non-purified test sample to said test compound. Preferably the compound is a test compound as defined elsewhere herein. Preferably the compound is a drug, e.g. an anticancer agent. Preferably the method is a high-throughput (drug) screen. These methods can be alternatively viewed as methods of assessing the ability of a test compound to inhibit apoptosis of cells in a non-purified test sample. The biomarker panel described herein advantageously provides the ability to investigate apoptotic processes in intact cells; the number and identities of biomarker proteins shifting in the assay, optionally over different time points post-induction of apoptosis, allows for investigation of apoptotic processes, e.g. signalling cascades, in the cells. Thus, the present invention also provides a method of investigating apoptotic processes in a population of cells, the method comprising the steps of: heating a non-purified test sample comprising said population of cells; and subsequently determining the level of soluble biomarker proteins in said heated test sample or in a soluble protein fraction thereof, or determining the level of insoluble biomarker proteins in an insoluble protein fraction of said heated test sample, comparing the determined level of each of said biomarker proteins with a control levels of said biomarker protein, and identifying the biomarker proteins for which there is a difference between the determined level and the control level, wherein said biomarker proteins comprise at least two proteins selected from the group (or panel) consisting of ADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CNOT2, COA7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HXRXPAO, HNRNPA3, HNRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, NOP 14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8. The panels and subsets of use in this method of the invention may be as defined elsewhere herein. Also provided is a method of monitoring the effectiveness of a therapeutic treatment by detecting apoptosis in a population of cells in a non-purified test sample obtained from a subject, wherein said subject has received said therapeutic treatment, wherein the method comprises the steps of any of the methods of the invention described elsewhere herein, wherein said control level is determined in a control sample obtained from said subject at an earlier time point in the course of therapeutic treatment. The level of apoptosis detected is indicative of the effectiveness of the treatment. Preferably the sample is a patient sample, preferably a cancer patient wherein said cells in said non-purified sample are cancer cells. The sample may be a tumour sample, e.g. biopsy. Preferably the therapeutic treatment is a chemotherapy or radiotherapy treatment. An increase in the level of apoptosis indicates an increased effectiveness of treatment, conversely a decrease in the level of apoptosis indicates a decreased effectiveness of treatment. The level (i.e. extent) of apoptotic status may correlate with the number of biomarker proteins for which there is detected a difference between the determined level of the soluble and / or insoluble biomarker protein as compared to a control level. Alternatively, the extent of apoptotic status (or index) may correlate with the extent (magnitude) of the difference between the determined level(s) of the soluble and / or insoluble biomarker protein(s) as compared to control level(s). The extent of the difference may be determined on an individual protein basis, or on the basis of the sum of the differences in the levels of multiple soluble or insoluble biomarker proteins as compared to control levels. Chemotherapy and radiotherapy treatment often induce high levels of apoptosis. However, neoplastic cells may be resistant to treatment. For example, in the case of leukaemia, particularly acute leukaemias, failure of malignant cells to undergo cell death in response to chemotherapy, is a major cause of treatment failure. As such, methods that are capable of detecting drug or radiotherapy-resistant cells, e.g. cancer cells, as well as a means to evaluate treatment effectiveness, would be desirable. The present invention provides such methods. Thus, in a further aspect, the present invention provides a method of identifying cells resistant to a therapeutic treatment, wherein the method comprises the steps of: heating a non-purified test sample comprising a population of cells that has been exposed to said therapeutic treatment; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, or determining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample, wherein resistance to said therapeutic treatment is identified if apoptosis is not detected, or if an insignificant amount of apoptosis is detected, i.e. if the increase between the level of the soluble biomarker protein as compared to a control level is insignificant (or the level is decreased), and / or wherein the decrease in the level of the insoluble biomarker protein as compared to a control level is insignificant (or the level is increased), wherein the control level has been obtained with a control sample which has not been exposed to said therapeutic treatment, and wherein said biomarker protein is as defined anywhere else herein. These methods may comprise a step before the heating step of exposing the non-purified test sample (and in embodiments, the control sample) to said treatment, or administering said treatment to said patient from which the patient sample has been obtained. Preferably the treatment is chemotherapy or radiotherapy. Preferable the sample is a subject or patient sample, preferably a cancer patient wherein said cells in said non-purified sample are cancer cells. The sample may be a tumour sample, e.g. biopsy. Also provided is a method of diagnosing an apoptosis-associated condition or disease in a subject. Preferably the subject is a mammal, preferably a human. The method comprises performing a method of the invention for detecting apoptosis, wherein the non-purified sample is a sample obtained from the subject, and wherein the condition or disease is diagnosed by the detection of apoptosis. Apoptosis-associated conditions are those characterised by or caused by aberrant apoptosis, e.g. excessive apoptosis, and the disease may be selected from the group consisting of chronic neurodegenerative disease, cancer, sepsis, trauma, hypoxia, anoxia, ischemia, spinal trauma, head trauma, lesions (wounds), and toxin-exposure. The invention will now be further described in the following non-limiting Examples in which: Figure 1 shows a cluster of nuclear proteins shifting in thermal stability (as determined by CETSA) 1 hour after apoptosis induction by venetoclax, including some known effector caspase targets (PARP1, LMNB1, LMNB2 PRKDC and MATR3), and some proteins previously not known to be early targets downstream of apoptosis activation (U2SURP, XRCC5, XRCC6, ZMYND8). String network and primary cellular localization of the BCL-2 inhibitor Venetoclax hit proteins at Ih drug exposure are shown. Mass-spectrometry was performed using isobaric-labeling and 3 biological replicates for each condition to minimize measurement errors. Figure 2 shows proteins shifting in thermal stability (as determined by CETSA) after venetoclax 6h (right-hand bars) and after the IAP inhibitor AT-IAP 4h (left-hand bars). Many proteins have not previously been considered to be affected downstream of apoptosis activation Figure 3 shows that XRCC5 shifts are downstream of caspase activation for venetoclax. WB-CETSA of XRCC5 at Ih and 6h with venetoclax and the pan-caspase inhibitor Z-VAD-FMK. Figure 4 shows via an orthogonal assay that apoptosis is induced at Ih and 6h Venetoclax treatment. Annexin V flow cytometry of venetoclax and pan-caspase inhibitor z-VAD-FMK. Data presented as percentage of cells (mean±SEM, n=5). Figure 5 shows that a selected set of CETSA apoptosis biomarker proteins shift (significant increase in soluble level of the protein vs. control) for a set of drugs known to induce apoptosis at the time points studied (Venetoclax, Azacytidine, Venetoclax / Azacytidine, AT-IAP and Paclitaxel). The combination of Venetoclax and Azacytidine show an increased apoptosis, i.e. demonstrating that greater shifts correlate with increased apoptotic status. To the right, changes on total protein in the cell using quantitate proteomics (QP) is shown for three proteins. This illustrates that very significant shifts can be seen in the CETSA experiment analysing the soluble fraction at different temperatures (ranging from 37C to 59C) while no significant difference are seen when quantifying total protein with QP. Data of CETSA experiment and QP are presented as FC log2 compared to control (Vehicle (0)), (n=3) ±SEM. EXAMPLES Example 1 The cells used in a first study, Kasumi-1 AML cells, over-express Bcl-2 family proteins, which are often overexpressed in cancer cells. Bcl-2 family proteins block the intrinsic apoptosis pathways by engaging pro-apoptotic BH3-only proteins. Apoptosis was induced by administering venetoclax, a Bcl-2 inhibitor currently in clinical use for e.g., several hematological cancers. This drug rapidly induces apoptosis by attenuating the apoptosis blockades. CETSA experiments were performed using non-purified test (drug treated) and control (non-treated) Kasumi-1 AML cells. Drug treatment comprised administering 100 nM venetoclax to cells at 1 h and 6 h incubation times. In parallel, quantitative proteomics (QP) data (total protein) was obtained. Aliquots of the samples were taken, and after heating each aliquot to a different temperature in a series of temperatures, soluble and insoluble protein fractions were separated by centrifugation, and the soluble fractions of each heated sample aliquot were prepared for, and analysed by, mass spectrometry. The mass spectrometry data was analysed to determine the level of every soluble protein in the heated sample aliquots. For each protein, at each temperature point, the relative level of the protein in the drug-treated heated sample as compared to in the non-treated control sample was determined. The relative level was expressed as the log2 fold-change of each soluble protein in the heated test sample as compared to in the control sample. 1 hour after venetoclax treatment, a distinct CETSA response was seen consisting of 16 proteins “shifted” over threshold. The results are shown in Figure 1. For each protein, there is displayed a series of bars, each of which shows the relative level of the protein detected in the soluble fraction of the test sample heated to a particular temperature, as compared to the soluble fraction of the control sample heated to the same temperature. The bars shown are the log2 fold change of the protein level in the test sample vs. control sample. These 16 proteins displayed a statistically significant increased level in the soluble fraction of the heated test sample (apoptosis induced) as compared to control (no apoptosis) at at least one temperature point. This change in level is termed a “shift”, as these proteins have a shifted thermal stability in the test sample as compared to the control sample. These proteins were ME3, ME2, TFAM, COA7, BCAP31, CCDC37, LMNB1, LMNB2, MATR3, U2SURP, DDX21, PARP1, XRCC5, PPKDC, XRCC6, and ZMYND8. The present inventors have therefore determined that these 16 proteins are biomarkers of apoptosis in a CETSA assay. The present inventors also determined that the dominating ensemble of proteins shifting at 1 h venetoclax treatment are nuclear and notably, several proteins of the response are well-established targets of effector caspases, including the DNA-repair coordinator PARP1 and the nuclear lamins LMNB1 and LMNB2. Additional well-validated caspase targets shifting are MATR3, a scaffold protein in the interface of nuclear lamins and chromatin; DDX21, a regulator of RNA-polymerase 1 localized in the nucleolus; and BCAP31, a membrane-bound receptor chaperone primarily localized to the endoplasmic reticulum (ER) membrane (26). The present inventors also determined that shifting proteins included two subunits of the KU complex (XRCC5 and XRCC6), involved in the repair of double-stranded DNA breaks. By establishing a WB-CETSA assay for XRCC5 and XRCC6, which were not expected caspase targets or previously known participants in apoptosis pathways, the present inventors confirmed the shift to be attenuated by the pan-caspase inhibitor Z-VAD-FMK (Figure 3). Using orthogonal Annexin V and Propidium Iodide flow cytometry-based apoptosis assay, the present inventors confirmed apoptosis induction by venetoclax at 1 h and 6 h, and that Z-VAD-FMK attenuated this response (Fig 4). Thus, the present inventors determined that XRCC5 and XRCC6 function in the apoptotic process downstream of apoptosis-activated caspases. Example 2 The study performed in Example 1 was repeated but wherein the test sample was exposed to venetoclax for 6 hours. After 6 h venetoclax treatment, the cellular response was found to be more complex, with 183 proteins shifting (increased soluble level) over threshold (control) with most of the responding proteins cluster being in ensembles assigned to the nucleus, encompassing proteins of the nuclear lamina, nucleolus, chromatin, DNA-repair and splicing complexes. The response also includes a subset of mitochondrial proteins, most being localized in the mitochondrial intermembrane space (MIMS). Due to the complexity of this 183-protein response, the present inventors anticipated that this encompassed CETSA shifts of proteins in a mix of primary and secondary apoptosis related processes. To identify proteins of a primary CETSA apoptosis response which could serve as stringent CETSA-biomarkers for apoptosis, the present inventors repeated the study with a different apoptosis activator, a so-called Inhibitor of Apoptosis Protein (IAP) antagonist with expected target engagement of XIAP and cIAPl& 2, IAP proteins. IAP antagonists rapidly induce apoptosis primarily by regulating extrinsic apoptosis-inducing receptor signalling and inhibition of caspases. Experiments with AT-IAP were performed in M0LM16 cells, in TNFa background (the latter to activate signalling pathways inhibited by IAPs). Again, the test sample was drug-treated (4 hour exposure), and control samples were not. Again, aliquots of the samples were taken, and after heating each aliquot to a different temperature in a series of temperatures, soluble and insoluble protein fractions were separated by centrifugation, and the soluble fractions of each heated sample aliquot were prepared for, and analysed by, mass spectrometry. The mass spectrometry data was analysed to determine the level of every soluble protein in the heated sample aliquots. For each protein, at each temperature point, the relative level of the protein in the drug-treated heated sample as compared to in the non-treated control sample was determined. The relative level was expressed as the log2 fold-change of each soluble protein in the heated test sample as compared to in the control sample. After 4h AT-IAP treatment, 148 proteins shifted over threshold (i.e. vs control). As for venetoclax most of the shifting proteins were in the nuclear lamina, nucleolus, chromatin, DNA-repair and splicing complexes, with 49 hits in common with venetoclax. The CETSA shifts for the 49 commonly shifted proteins are shown in Figure 2. For each protein, there is displayed two series of bars, one obtained from venetoclax treated samples (lefthand series) and one obtained from AT-IAP treated samples (right-hand series). Within each series, each of the bars shows the relative level of the protein detected in the soluble fraction of the drug-treated test sample heated to a particular temperature, as compared to the soluble fraction of the control sample heated to the same temperature. The bars shown are the log2 fold change of the protein level in the test sample vs. control sample. These 49 proteins shown in Figure 2 are therefore identified as biomarkers of apoptosis in CETSA assays, as demonstrated by the present inventors using entirely different initiators of apoptosis. The present inventors noted that of the 16 proteins shifting in thermal stability 1 hour after venetoclax treatment, twelve were observed to have shifting in thermal stability at the 6 hour time point: TFAM, BCAP31, LMNB1, LMNB2, PARP1, PPKDC, MATR3, XRCC5, XRCC6, ZMYND8, U2SURP, and DDX2. This demonstrates that the method is advantageously useful, allowing for analysis of the apoptotic process at different time points. Very early stage apoptosis events (i.e. initiation or activation) are observable (Figure 1), with subsequent early-stage protein processes, e.g. more comprehensive modulations of the nucleolus, chromatin and splicing complex proteins, also being detectable. Such prominent modulations of nuclear proteins in apoptosis initiation and early-stage apoptosis has not previously been demonstrated, and most of the biomarker proteins identified have not previously been shown downstream of apoptosis pathway activation. Example 3 The study performed in Examples 1 and 2 was repeated with four further apoptosisinducing drug treatments of the test sample: i) Azacytidine (6 hour exposure), ii) Venetoclax+Azacytidine (6 hour exposure), iii) methylene quinuclidinone (MQ) (6 hour exposure) and iv) Paclitaxel (8 hour exposure). The results obtained with the biomarker proteins CDC5L, CHD4, DDX21, FTSJ3, HNRNPAO, HNRNPA3, LMNB1, LMNB2, MATR3, M0RC3, NOP14, PARP1, RBMX, SMARCC1, U2SURP, XRCC5, XRCC6, ZC3H13, and ZMYND8 are shown in Figure 5. All biomarker proteins demonstrated a significant shift (increased soluble level vs. control) as compared to control, at at least one temperature point, which demonstrates their utility as biomarkers of apoptosis. The combination of Venetoclax and Azacytidine treatment show particularly high protein shifts, i.e. increased apoptosis. This demonstrates that greater shifts correlate with increased apoptotic status. To the right of each plot, changes in total protein in the cell using quantitate proteomics (QP) is shown for three proteins. This illustrates that very significant shifts can be seen in the CETSA experiment analysing the soluble fraction at different temperatures (ranging from 37C to 59C) while no significant difference are seen when quantifying total protein with QP. Additionally, the assay was repeated with two non apoptosis inducing drugs (Cytarabine (6 hour exposure) and Decitabine (6 hour exposure)) and protein shifts were not observed in any of the biomarker proteins (data not shown). Materials and Methods Cell Lines Several cancer cell lines of human origin were purchased from ATCC: MCF7 (HTB-22, derived from pleural effusion at the metastatic site of a 69-year-old female with breast adenocarcinoma), Kasumi-1 (CRL-2724) established from the peripheral blood of an acute myeloid leukemia (AML) patient, and OVCAR-3 (HTB-161) isolated from the malignant ascites of a patient with progressive adenocarcinoma of the ovary. Human acute myeloid leukaemia cell line M0LM16 was from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures. MCF7 cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) high glucose medium (D6429, Sigma) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (10500-064, Gibco), Kasumi-1 were maintained in RPML1640 medium (R8758, Sigma) supplemented with 20% FBS, OVCAR-3 were maintained in RPMI-1640 medium (R8758, Sigma) supplemented with 20% FBS and M0LM16 were maintained in RPMI-1640 medium (R8758, Sigma) supplemented with 10% FBS. For cell passaging, MCF7 and OVCAR-3 cells were washed with Hanks' Balanced Salt Solution (HBSS) -CaCW-MgCh (14175-053, Gibco) and detached by 3 minutes incubation with TrypLE Select lx (12563-029, Gibco) at 37°C. Drugs used to induce apoptosis Paclitaxel (SI 150, Selleck Chern), Z-VAD-FMK (S7023, Selleck Chern), methylene quinuclidinone (MQ) (provided by Aprea Therapeutics) and staurosporine (62996, ThermoFisher Scientific) were solubilised at 50 mM in DMSO. Venetoclax (S8048, Selleck Chern) and AT-IAP (provided by Astex Pharmaceuticals) were solubilised at 10 mM in DMSO. All compound stocks were aliquoted and stored at -80°C. The cellular thermal shift assay (CETSA) Non-purified samples were used. Cells were seeded in the corresponding complete medium at a suitable cell density (previously determined). The cells were then treated with either drug (test sample) or vehicle (control sample) diluted in the corresponding growth medium, and incubated at 37°C and 5% CO2. Adherent cells were briefly washed in HBSS ^CaCW-MgCh, detached from the flask using TrypLE, collected in HBSS ~CaCh / +MgCh, pelleted for 3 min at 300*g, and resuspended in HBSS -CaChZ+MgCh. Suspension cells were pelleted for 3 min at 300*g, washed twice in HBSS -CaChZ-MgCh, and resuspended in HBSS +CaC12 / +MgC12. Cells corresponding to each treatment condition were aliquoted into PCR tubes then heated in a Veriti thermal cycler (Applied Biosystems) for 3 min at the selected temperatures (between 37-59°C, Table 1). After heating, all samples were subjected to 3 cycles of freezethawing in liquid nitrogen in the presence of lx Halt™ protease inhibitor cocktail (1861279, Thermo Scientific) or 1 Halt™ Protease and Phosphatase Inhibitor Cocktail (1861284, Thermo Scientific). The soluble protein fraction was isolated by 20 min centrifugation at 20 OOOxg at 4°C, and analysed using either LC-MS or western blotting. For QP samples, 8 M Urea (freshly prepared) with lx Halt™ Protease and Phosphatase Inhibitor Cocktail (1861284, Thermo Scientific) was added to samples, vortexed and incubated for 15 minutes at room temperature and then sonicated. Table 1 Cell line Temperatures Drug Concentration Time point MOLM-16 37C / 47C / 50C / 52C / 54C / 57C AT-IAP 100 nM 4h TNFa 1 ng / mL Kasumi-1 QP / 37C / 47C / 50C / 52C / 54C / 57C Venetoclax 100 nM Ih 6h MCF-7 37C / 47C / 50C / 53C / 55C / 59C Paclitaxel 25 pM 8h OVCAR-3 37C / 42C / 46C / 52C / 58C MQ 6 pM 6h Kasumi-1 QP / 37C / 47C / 50C / 52C / 54C / 57C Azacitidine 5 pM 6h Kasumi-1 QP / 37C / 47C / 50C / 52C / 54C / 57C Venetoclax + Azacytidine 100 nM / 5 pM 6h Western blot Western blotting was performed on protein extracts obtained by freeze-thawing. 1x Halt™ Protease and Phosphatase Inhibitor Cocktail (1861284, Thermo Scientific) was added to samples. Protein extract samples were mixed with NuPAGE loading buffer consisting of NuPAGE LDS sample buffer (NP0008, Life technologies) and reducing agent (NP0009, Life Technologies). Proteins were separated on NuPAGE 4-12% Bis-Tris midi gels (WG1403BX10, Invitrogen) for 45-55 min at 200 V. Separated proteins were transferred to nitrocellulose membranes using the iBlot2 system (Invitrogen) and iBlot 2 NC Regular Stacks (IB23001, Invitrogen). Membranes were blocked in 5% (w / v) non-fat milk (Semper AB) in TBS with 0.05% Tween 20 (Medicago 09-7510-100) (TBS-T) for 1 h with gentle shaking. Incubation with primary antibody was performed overnight at 4°C and with gentle shaking. After washing in TBS-T for 3x 10 min, the membranes were incubated with secondary antibodies for 1 h, washed again 3x10 min in TBS-T and developed using either Clarity™ Western ECL Substrate (170-5061, BioRad) or Clarity Max™ Western ECL Substrate (1705062, BioRad). All antibodies were prepared in 5% (w / v) non-fat milk in TBS-T at the dilutions listed below. The chemiluminescent signal was detected using the ChemiDoc™ XRS+ imaging system from BioRad and the band intensities were quantified using ImageLab™ software (BioRad). Primary antibodies: 1:1000 anti-cleaved PARP (Asp214) (9541, Cell Signaling), 1:1000 anti-LMNB1 N-term (12586, Cell Signaling), anti-MATR3 (ab 151714, Abeam), 1:10000 anti-XRCC5 (ab 109012, Abeam). Sample preparation for LC-MS Protein concentrations were determined in the samples using the DC protein assay (500-0116, BioRad) according to the manufacturer’s protocol and equal amounts of total protein from each condition were used for sample preparation. Samples were dried and resuspended in 50 mM triethylammonium bicarbonate (TEAB) buffer (T7408, Sigma) in H2O (LC-MS grade, 115333, Merck), reduced with 5 mM TCEP (77720, Bond-breaker™, Thermo Scientific) at 65°C for 30 min, followed by alkylation with 15mM 2-chloroacetamide (CAA) (C0267, Sigma) at 37°C in the dark. Digestion of the samples was performed with 1:50 Lys-C (Wako Chemicals Ltd) at 37°C for 2h, followed by 1:50 SOLu-Trypsin (EMS0004, Sigma) at 37°C overnight, for QP samples, before LysC addition amples were diluted to 2M UREA with lOOmM TEAB and before Trypsin addition they were diluted to IM UREA also in lOOmM TEAB. After digestion efficiency was checked, the peptides were labelled with Isobaric Tandem Mass Tags (TMT)-lOplex (90110, Thermo Scientific) or (TMTpro)-16plex (A44522, Thermo Scientific) at 37°C for 3h, followed by checking labelling efficiency (>95% TMT-labelled PSMs). Three biological replicates from the treatment conditions at the same temperature were labelled as a set. The labelled samples from the same TMT set were pooled and the labelling reaction was stopped by adding 10% trifluoroacetic acid (Sigma) to reach pH <3. Samples were dried in a Speed Vac vacuum concentrator (Thermo Scientific) and desalted using Oasis HLB lec (lOmg) extraction cartridges (186000383, Waters) according to manufacturer’s protocol. Offline pre-fractionation of the samples was performed by high pH reverse-phase liquid chromatography using the using a Xbridge Peptide BEH C18, 300 A, 3.5 pm, 2.1 mm x 250 mm column (#186003610, Waters) The fractions were concatenated and dried using a SpeedVac vacuum concentrator (Thermo Scientific). LC-MS The digested, labeled, and dried peptide sample fractions were resuspended in 0.1% FA (LC-MS grade, 533002, Merck) in H2O (LC-MS grade, 115333 Merck). Online chromatography was performed using Dionex UltiMate 3000 UPLC system coupled to a Q Exactive mass spectrometer (Thermo Scientific). Each fraction was separated on a 50 cm x 75 pm (ID) EASY-Spray analytical column (Thermo Scientific) in a gradient of programmed mixture of solvent A (0.1% formic acid in H2O) and solvent B (99.9% acetonitrile, 0.1% formic acid). MS data were acquired using a top 12 data-dependent acquisition method. Full scan MS spectra were acquired in the range of 375-1500 m / z at a resolution of 60,000 and AGC target of 3e6; Top 12 dd-MS2 60,000 and 3e6 with isolation window at 1.2 m / z. Protein identification and quantification in MS data Protein identification was performed by Proteome Discoverer software (Thermo Scientific), using Sequest HT (Thermo Scientific) search engines to search against reviewed human Uniprot databases (downloaded on 13 Jan 2017, including 42105 sequence entries and another downloaded on 23 Jul 2018, including 9606 sequence entries). MS precursor mass tolerance was set at lOppm, fragment mass tolerance 0.02 Da, and maximum missed cleavage sites of 2. Dynamic modifications searched for Oxidation (M), Deamidation (NQ), and Acetyl N-terminal protein. Static modifications: Carbamidomethyl (C) and TMTIOplex or TMTpro-16plex (K and peptide N terminus). Only the spectrum peaks with signal-to-noise ratio (S / N) >4 were chosen for searches. The false discovery rate (FDR) was set to 1% at both PSM and protein group levels. Only the unique and razor peptides were used for protein assignment and abundance quantification. Isotopic correction of the reporter ions in each TMT channel was performed according to the product sheet. Only the master proteins in the protein group were used for downstream analysis. Quantitative MS data analysis and visualization Quantified protein / peptide abundances were imported into the R environment (http: / / www.R-project.org / ) to facilitate the data analysis and visualization. Data cleaning, normalization, and calculations of protein abundance and thermal stability differences in each condition were performed). Strict criteria for hit selection were applied for all datasets. It was required that proteins should be quantified by at least 2 peptide spectrum matches (PSMs). As cut-off criteria an absolute mean fold change and a standard error of the mean (SEM)-scale factor were used as cut-off. A mean log2 fold change cut-off >0.25 or 0.2 and ± 4* SEM as compared to reference condition was applied. Hits with only 1 temperature over threshold were further inspected where some hits with e.g. high SEMs on most other temperatures were manually deleted. Data Analysis and visualisation All graphs were generated using GraphPad Prism, R environment, cytoscape or Biorender. All data are presented as mean with error bars representing the standard error of the mean (SEM). Error bars that are smaller than the displayed data points are not displayed by the software. Details regarding replicates for each experiment can be found in the figure legends. Sigmoidal curves were fit (where appropriate) using R environment. Unpaired t-tests were performed using GraphPad Prism and the results are displayed in figures and figure legends. The flow cytometry data was analyzed on FlowJo vl0.8 and GraphPad Prism was used to represent the data. Annexin V assay Assay was performed according to the provided protocol by ApoDETECT™ Annexin V-FITC kit (Invitrogen Cat#: 33-1200). The treated cells along with controls for all time points were harvested together at the same time. Cells were spun down for 1 min at 3000 rpm. Pelleted cells were washed in ice cold PBS (pH 7.4) while gently resuspending the pellet. Cells were centrifuged again for 1 min at 3000 rpm. Resuspended pelleted cells in IX binding buffer and cell density was adjusted to 5 x 105 cells / ml. 10 pl of Annexin V-FITC was added to 190 pl of cell suspension. Incubate 10 min at room temperature. Cells were washed with IX binding buffer, spun down at 3000 rpm for 1 min and pellet resuspended in 190 pl binding buffer. 10 pl of 20 mg / ml propidium iodide solution added. Cells were analyzed by flow cytometry on Novocyte 3000.

Claims

1. A method for detecting apoptosis, wherein the method comprises the steps of: heating a non-purified test sample comprising a population of cells; and subsequently determining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, ordetermining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample,wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level,and wherein said biomarker protein is selected from the group consisting ofADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CN0T2, C0A7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPAO, HNRNPA3, HNRNPC, HNRNPU, IM MT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, N0P14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8.

2. The method of claim 1, wherein the method comprises the steps of:a) heating a non-purified test sample comprising a population of cells;b) separating soluble from insoluble protein in the product of step a) to produce soluble and insoluble protein fractions of said heated test sample; andc) determining the level of the biomarker protein in either or both of said soluble and insoluble protein fractions of said heated test sample, wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in said soluble protein fraction of said heated test sample as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in the insoluble protein fraction of said heated test sample as compared to a control level.

3. The method of claim 2, wherein step b) is a step of centrifugation or filtration.

4. The method of any one of claims 1 to 3, wherein the level of the biomarker protein isdetermined by affinity binding or by mass spectrometry.

5. The method of any one of claims 1 to 4, wherein the method comprises the steps of:a) heating a non-purified test sample comprising said population of cells;b) separating soluble from insoluble protein in the product of step a) to produce soluble and insoluble protein fractions of said heated test sample; andc) subjecting either or both the soluble and insoluble protein fractions of step b) to mass spectrometry analysis to produce mass spectrometry data, and analysing the mass spectrometry data to determine the level of the biomarker protein,wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in the soluble protein fraction of said heated test sample as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in the insoluble protein fraction of said heated test sample as compared to a control level.

6. The method of any one of claims 1 to 5, wherein the method is repeated using one or more different temperatures in said step of heating, and wherein apoptosis is determined by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level, after heating to at least one temperature.

7. The method of any one of claims 1 to 6, wherein the heating step comprisesi) heating the non-purified test sample to a series of temperatures in a temperature range and obtaining an aliquot of the sample after heating at each temperature, orii) obtaining a plurality of non-purified test sample aliquots and heating each sample aliquot to a different temperature within a series of temperatures in a temperature range,to obtain a plurality of heated test sample aliquots;wherein the subsequent steps of the method are performed on each of said heated test sample aliquots,and wherein apoptosis is detected by an increase in the level of the soluble biomarker protein in said heated test sample aliquot or soluble protein fraction thereof as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein in the insoluble protein fraction of said heated test sample as compared to a control level.

8. The method of any one of claims 1 to 7, wherein the level of the biomarker protein is determined in the soluble protein fraction of said heated test sample or aliquot thereof.

9. The method of any one of claims 1 to 8, wherein said biomarker protein is selected from the group consisting of CDC5L, CHD4, DDX21, FTSJ3, HNRNPAO, HNRNPA3, LMNB1, LMNB2, MATR3, M0RC3, N0P14, PARP1, RBMX, SMARCC1, U2SURP, XRCC5, XRCC6, ZC3H13, and ZMYND8.

10. The method of any one of claim 1 to 9, wherein the method comprises determining the soluble and / or insoluble level of at least three of said biomarker proteins, and wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level, for each of said at least three biomarker proteins.

11. The method of any one of claims 1 to 10, wherein the method comprises determining the soluble and / or insoluble level of at least ten of said biomarker proteins, and wherein apoptosis is detected by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level, for at least five of said biomarker proteins.

12. The method of any one of claims 1 to 11, wherein said increase and / or decrease is statistically significant.

13. The method of any one of claims 1 to 12, wherein said apoptosis is nuclear apoptosis, preferably early nuclear apoptosis.

14. The method any one of claims I to 13, wherein said step of heating comprises heating to a series of temperatures in the temperature range of 37°C to 60°C.

15. The method of any one of claims 1 to 14, wherein said control level is the level of said soluble or insoluble biomarker protein in an equivalent heated non-purified control sample or sample aliquot, or protein fraction thereof.

16. The method of any one claims 1 to 15, wherein said non-purified sample is a liquid culture of cells, a sample derived from a plant or part thereof, or a sample obtained from a human or animal subject.

17. The method of any one of claims 1 to 16, wherein the non-purified sample is subjected to said heating step before or after a step of lysis is performed, preferably before a step of lysis is performed.

18. A method for identifying an apoptosis-inducing agent, wherein the method comprises the steps of:heating a non-purified test sample comprising a population of cells that has been exposed to a test compound; and subsequentlydetermining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, ordetermining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample,wherein the test compound is identified as an apoptosis inducing agent by an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level,and wherein said biomarker protein is selected from the group consisting ofADNP, AIFM1, BCAP31, BCLAFI, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CN0T2, C0A7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPAO, HNRNPA3, HNRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, N0P14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8.

19. A method of investigating apoptotic processes in a population of cells, the method comprising the steps of:heating a non-purified test sample comprising said population of cells; and subsequently determining the level of soluble biomarker proteins in said heated test sample or in a soluble protein fraction thereof, ordetermining the level of insoluble biomarker proteins in an insoluble protein fraction of said heated test sample,comparing the determined level of each of said biomarker proteins with a control levels of said biomarker protein,and identifying the biomarker proteins for which there is a difference between the determined level and the control level,wherein said biomarker proteins comprise at least two proteins selected from the group (or panel) consisting of ADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CN0T2, C0A7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPAO, HNRNPA3, HNRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, NOP14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8.

20. A method of diagnosing an apoptosis-associated condition or disease in a subject, wherein the method comprises the steps of:heating a non-purified test sample comprising a population of cells obtained from said subject; and subsequentlydetermining the level of a soluble biomarker protein in said heated test sample or in a soluble protein fraction thereof, ordetermining the level of an insoluble biomarker protein in an insoluble protein fraction of said heated test sample,wherein a diagnosis is made by determining an increase in the level of the soluble biomarker protein as compared to a control level, and / or by a decrease in the level of the insoluble biomarker protein as compared to a control level,and wherein said biomarker protein is selected from the group consisting ofADNP, AIFM1, BCAP31, BCLAF1, CBX1, CBX5, CCDC37, CDC5L, CHCHD3, CHD4, CNOT2, COA7, DDX18, DDX21, ESF1, FTSJ3, GABPB1, GNL2, HNRNPAO, HNRNPA3, HNRNPC, HNRNPU, IMMT, LMNA, LMNB1, LMNB2, LUC7L3, MATR3, ME2, ME3, M0RC3, NCL, NOP14, PARP1, PPM1G, PRKDC, RBM5, RBMX, RRP1, RSF1, SMARCC1, TFAM, TOPI, U2SURP, UBR7, UBTF, WDHD1, XRCC5, XRCC6, ZC3H13, and ZMYND8.

Images