A method, apparatus, system, and medium for predicting drug hepatotoxicity

Abstract

The application provides a drug hepatotoxicity prediction method, device, system and medium, the method comprising: sequencing human liver cells after being contacted with a drug to be predicted to obtain a genome expression profile of the human liver cells after being contacted with the drug to be predicted; calculating the sum of the absolute values of the fold changes of the genes in the current gene pathway as the toxicity score of the current gene pathway according to the expression profile; calculating the Z value of the current gene pathway according to the toxicity score of the current gene pathway, the average value and the standard deviation value of the toxicity scores of the gene pathways calculated multiple times by a plurality of preset gene pathways; and predicting that the drug to be predicted has hepatotoxicity when the Z value is greater than 1.5. Therefore, the application can accurately, timely, effectively and sensitively predict drug hepatotoxicity without a large number of living animals as an experimental basis, and has good advantages in safety, environmental protection and animal protection. The application has the advantages of short experimental period and low cost.

Description

Technical Field

[0001] This application relates to the field of biomedicine, and in particular to a method, apparatus, system and medium for predicting drug hepatotoxicity. Background Technology

[0002] Drug safety is a key factor determining the success or failure of innovative drug development. Preclinical safety evaluation is an important basis for determining the safety performance indicators of new drugs and reducing their toxic effects. Predicting drug hepatotoxicity in the early stages of drug development and during clinical use is of great significance for improving the success rate of development and promoting rational drug use.

[0003] Traditional methods for predicting drug hepatotoxicity primarily rely on in vitro and in vivo experimental models to assess drug toxicity and safety. These methods require extensive live animal studies, resulting in lengthy and costly experimental cycles. With increasingly stringent regulations regarding safety, environmental protection, and animal welfare, and with global market competition demanding significantly shorter drug development cycles, improving the accuracy and applicability of drug hepatotoxicity prediction remains a significant challenge due to the complexity of its mechanisms. Therefore, overcoming the shortcomings of traditional methods and developing efficient drug hepatotoxicity prediction techniques is of great importance. Summary of the Invention

[0004] In view of this, a summary section is provided to briefly introduce the concepts, which will be described in detail in the detailed description section below. This summary section is not intended to identify key or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] The purpose of this application is to provide a method, device, system, and medium for predicting drug hepatotoxicity, which can accurately and effectively predict drug hepatotoxicity and has advantages in terms of safety, environmental protection, and animal protection. It also features a short experimental cycle and low cost.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] In a first aspect, embodiments of this application provide a method for predicting drug hepatotoxicity, including:

[0008] Human liver cells exposed to the drug to be predicted were sequenced to obtain the genomic expression profile of the human liver cells after exposure to the drug to be predicted.

[0009] Based on the expression profile, the sum of the absolute values ​​of the fold changes of genes contained in the current gene pathway is calculated as the toxicity score of the current gene pathway;

[0010] The Z-value of the current gene pathway is calculated based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times.

[0011] When the Z value is greater than 1.5, the drug to be predicted is predicted to have hepatotoxicity.

[0012] In one possible implementation, calculating the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times includes:

[0013] ;

[0014] Where Z is the Z-value, X is the toxicity score of the current gene pathway, μ0 is the mean, and σ is the standard deviation.

[0015] In one possible implementation, the current gene pathway includes: cell cycle gene pathway, replication gene pathway, chemical oncogenic gene pathway, apoptosis gene pathway, lysosomal gene pathway, necroptosis gene pathway, p53 signaling gene pathway, cellular senescence gene pathway, bile duct hyperplasia gene pathway, liver fibrosis gene pathway, liver steatosis gene pathway, cell division regulation gene pathway, hematopoietic regulation gene pathway, glycerophospholipid synthesis gene pathway, cell necrosis gene pathway, or leukocyte infiltration gene pathway.

[0016] In one possible implementation, the sequencing includes high-throughput sequencing.

[0017] Secondly, embodiments of this application provide a drug hepatotoxicity prediction device, comprising:

[0018] A sequencing unit is used to sequence human liver cells exposed to the drug to be predicted, and to obtain the genomic expression profile of the human liver cells exposed to the drug to be predicted.

[0019] The toxicity score calculation unit is used to calculate the sum of the absolute values ​​of the fold differences of genes contained in the current gene pathway based on the expression profile, and use this sum as the toxicity score of the current gene pathway.

[0020] The Z-value calculation unit is used to calculate the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times.

[0021] The prediction unit is used to predict that the drug to be predicted has hepatotoxicity when the Z value is greater than 1.5.

[0022] In one possible implementation, the Z-value calculation unit is specifically used to calculate the Z-value of the current gene pathway:

[0023] ;

[0024] Where Z is the Z-value, X is the toxicity score of the current gene pathway, μ0 is the mean, and σ is the standard deviation.

[0025] In one possible implementation, the current gene pathway includes: cell cycle gene pathway, replication gene pathway, chemical oncogenic gene pathway, apoptosis gene pathway, lysosomal gene pathway, necroptosis gene pathway, p53 signaling gene pathway, cellular senescence gene pathway, bile duct hyperplasia gene pathway, liver fibrosis gene pathway, liver steatosis gene pathway, cell division regulation gene pathway, hematopoietic regulation gene pathway, glycerophospholipid synthesis gene pathway, cell necrosis gene pathway, or leukocyte infiltration gene pathway.

[0026] In one possible implementation, the sequencing includes high-throughput sequencing.

[0027] Thirdly, embodiments of this application provide a drug hepatotoxicity prediction system, including:

[0028] Memory, used to store computer programs;

[0029] A processor is used to implement the steps of the drug hepatotoxicity prediction method as described above when executing the computer program.

[0030] Fourthly, embodiments of this application provide a computer-readable medium storing a computer program, which, when processed and executed, implements the steps of the drug hepatotoxicity prediction method described above.

[0031] Compared with the prior art, the embodiments of this application have the following beneficial effects:

[0032] This application provides a method, device, system, and medium for predicting drug hepatotoxicity. The method includes: sequencing human liver cells exposed to the drug to be predicted to obtain a genomic expression profile of the liver cells; calculating the sum of the absolute values ​​of the fold changes of genes in the current gene pathway based on the expression profile, as the toxicity score of the current gene pathway; calculating the Z-score of the current gene pathway based on the toxicity score of the current gene pathway, the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times; and predicting that the drug to be predicted has hepatotoxicity when the Z-score is greater than 1.5. Therefore, this application can accurately, effectively, and sensitively predict drug hepatotoxicity without requiring a large number of live animals as an experimental basis, offering advantages in terms of safety, environmental protection, and animal welfare. It also has advantages such as a short experimental cycle and low cost. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0035] Figure 1 A flowchart of a drug hepatotoxicity prediction method provided in an embodiment of this application is shown;

[0036] Figure 2 This paper presents a thermodynamic diagram showing the Z-value calculated when the drug to be predicted is aristolochic acid lactam II, as provided in an embodiment of this application.

[0037] Figure 3 This paper presents a thermodynamic diagram showing the Z-value calculated when the drug to be predicted is blister beetle, cocklebur, or atractylodes lancea, as provided in the embodiments of this application.

[0038] Figure 4 A schematic diagram of a drug hepatotoxicity prediction device provided in an embodiment of this application is shown. Detailed Implementation

[0039] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the specific embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0040] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0041] As described in the background section, the applicant's research has found that drug safety is a key factor determining the success or failure of innovative drug development, and preclinical safety evaluation is an important basis for determining the safety performance indicators of new drugs and reducing their toxic effects. Predicting drug hepatotoxicity in the early stages of drug development and during clinical use is of great significance for improving the success rate of research and development and promoting rational drug use.

[0042] Traditional methods for predicting drug hepatotoxicity primarily rely on in vitro and in vivo experimental models to assess drug safety and toxicity. These methods require a large number of live animals for experimentation, resulting in long experimental cycles and high costs. With increasingly stringent regulations regarding safety, environmental protection, and animal welfare, and with global market competition demanding significantly shorter drug development cycles, improving the accuracy and applicability of drug hepatotoxicity prediction remains a significant challenge due to the complexity of its mechanisms. Therefore, overcoming the shortcomings of traditional methods and developing efficient drug hepatotoxicity prediction techniques is of great importance.

[0043] Predicting drug hepatotoxicity in the early stages of drug development and clinical use is crucial for improving research success rates and promoting rational drug use. For example, in the research of traditional Chinese medicine (TCM), as the application scope of TCM becomes increasingly broad, reports of toxicity or adverse reactions caused by TCM are frequent. The difficulty in TCM toxicity research lies in the complexity of its components and the unpredictability of drug interactions; its mechanisms of action and dose-response relationships are not fully elucidated. Drug-induced hepatotoxicity of TCM has become a focus of attention in academia and industry. Whether for the clinical application of TCM or the development of new drugs, early evaluation of TCM hepatotoxicity, identification of toxic chemical components, and exploration of their mechanisms of action are of great significance.

[0044] With the increasing application of high-throughput sequencing, toxicological omics (genomics, proteomics, and metabolomics), and bioinformatics in the fields of toxicology and food safety evaluation, establishing in vitro models in non-experimental animals, identifying and characterizing toxic pathways related to toxic outcomes, establishing genes or gene sets for effective prediction of hepatotoxicity, developing suitable algorithms, and constructing accurate and reliable toxicity prediction models remain extremely challenging tasks.

[0045] High-throughput screening, toxicological omics (genomics, proteomics and metabolomics), bioinformatics, systems biology and translational medicine are increasingly being applied to the fields of toxicology and food safety evaluation, which plays an important role in deepening our understanding of the potential mechanisms of action of toxic substances and predicting the adverse consequences of related toxic substances.

[0046] Therefore, identifying and characterizing toxic pathways related to toxic outcomes in vivo, revealing the molecular toxicity mechanisms of endogenous key hazards and their metabolites, and determining key toxic pathways and hazard biomarkers that can be used for safety risk assessment and toxicity detection are crucial for the innovation of food safety assessment technologies. Internationally, specialized databases such as "connectivity maps" (Lamb et al., 2006), DrugMatrix (Ganter et al., 2006), and TG-GATEs (Igarashi et al., 2015) link expression data with chemically induced bioperturbations and toxicity. The Broad Institute in the United States has established a connectivity map (CMap) for small molecule drugs treating human cells using gene expression profiling technology. This map can be used to identify certain properties of new chemical substances and to identify compounds with positive or negative associations by comparing it with reference expression profiles in a background library. This function can be used to effectively predict the toxicity of compounds. The Japanese Toxicogenomics Project has produced the large-scale toxicogenomics database Open TG-GATEs, which has established a method for evaluating the potential hepatotoxicity of drugs based on whole-genome expression profiling data. This database provides systematic and comprehensive toxicogenomics data related to the liver. These developments provide new insights into the development of new technologies for food safety assessment. Auerbach (2016) developed a robust, "universal" 32-gene marker for in vivo genotoxicity by mining expression data from a wide range of studies involving many tissues and different species. Researchers in biochemistry and molecular cell biology at Georgetown University School of Medicine developed an in vitro transcriptomic biomarker, TGx-DDI, capable of predicting with up to 90% accuracy whether candidate drugs or other compounds cause the type of cell damage in cancer.

[0047] However, current methods based on toxicogenomics are still based on changes in the expression of a single gene or a group of genes, and do not correspond to phenotypes of specific organ damage.

[0048] To address the aforementioned technical problems, this application provides a method, apparatus, system, and medium for predicting drug hepatotoxicity. The method includes: sequencing human liver cells exposed to the drug to be predicted to obtain a genomic expression profile of the liver cells; calculating the sum of the absolute values ​​of fold changes in genes within the current gene pathway based on the expression profile, as the toxicity score of the current gene pathway; calculating the Z-score of the current gene pathway based on the toxicity score of the current gene pathway, the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times; and predicting hepatotoxicity of the drug when the Z-score is greater than 1.5. Therefore, this application eliminates the need for a large number of live animals as experimental subjects, enabling accurate, timely, effective, and sensitive prediction of drug hepatotoxicity, offering advantages in safety, environmental protection, and animal welfare. It also features a short experimental cycle and low cost.

[0049] This method overcomes the limitations of traditional compound hepatotoxicity prediction methods, enabling early prediction of delayed hepatotoxicity of compounds through toxic pathways or gene sets. This provides an efficient, accurate, and rapid method for predicting the hepatotoxicity of compounds or single drugs in the process of new drug development, including toxicity and safety assessment and food safety.

[0050] See Figure 1 The diagram shown is a flowchart of a drug hepatotoxicity prediction method provided in an embodiment of this application, including:

[0051] S101: Sequencing human liver cells exposed to the drug to be predicted yields the genomic expression profile of the human liver cells exposed to the drug to be predicted.

[0052] In this embodiment of the application, human liver cells exposed to the drug to be predicted can be sequenced to obtain the genomic expression profile of the human liver cells after exposure to the drug to be predicted.

[0053] Specifically, the first step can be to prepare the drug to be predicted. The drug to be predicted can be, for example, a monomeric substance. The monomeric substance is prepared into a storage solution using an appropriate solvent, such as sterile water, PBS (Phosphate Buffer Saline), or DMSO (Dimethyl sulfoxide).

[0054] If the drug to be predicted is a traditional Chinese medicine (TCM), then the single herb slices or the TCM to be predicted will be extracted, and the extract will be used for cell treatment. Extraction methods can be used, but are not limited to, the following: For example, for alcohol extraction, the single herb slices can be crushed using a pulverizer and placed in a Soxhlet extractor. 90% ethanol can be added, and the extraction can be performed under reflux using a Soxhlet extractor. After filtration, the extract is concentrated at 45°C to obtain a paste, which is then freeze-dried for 24 hours to obtain the corresponding single herb extract. Alternatively, the extract can be pulverized into powder using a pulverizer. It can be soaked in 90% ethanol solution, ultrasonically extracted for 10-30 minutes, allowed to stand completely, and the supernatant concentrated at 70°C. The concentrated solution is then frozen at -80°C for at least 6 hours and then vacuum-dried to obtain a dried powder.

[0055] Then, human liver cells can be obtained. Specifically, take HepG2 (human hepatocellular carcinoma) culture flasks from the incubator, add trypsin to each flask, gently shake to ensure the entire cell surface is covered with trypsin, place in the incubator to digest until some cells float, add complete culture medium to stop digestion, use a Pasteur pipette to agitate the culture medium to detach the cells, separate them from the liver, discard the supernatant, add culture medium and gently agitate to mix the cells, count them using a cell counting chamber, and plate them in 96-well or 384-well cell culture plates. Incubate in an incubator for 24 hours.

[0056] Next, human liver cells were treated with the drug to be predicted. Specifically, the cell culture plate was removed from the incubator, and the culture medium was slowly discarded. The cells were washed once with PBS. Experimental wells and control wells were set up. The cells treated with the drug or extract to be predicted served as the treatment group, and the cells were treated with the solvent of the drug or extract to be predicted as the control group. Stock solutions of monomers or traditional Chinese medicine extracts at different concentration gradients were added to the experimental wells. The cells were incubated for 24 hours.

[0057] Once human liver cells exposed to the drug to be predicted are obtained, they can be sequenced to obtain the genomic expression profile of these cells. Specifically, the cell culture plate can be removed, the culture medium discarded, and PBS added for washing. Cell lysis buffer is added to each well, and the cells are gently pipetted five times to mix (operate on ice), and lysed on ice for 5 minutes. The lysed RNA is then reverse transcribed to produce cDNA. After fragmentation, end repair and completion, A-tailing, and adapter ligation, library enrichment is performed. The resulting library is then subjected to Illumina high-throughput sequencing.

[0058] In one possible implementation, the sequencing method provided in this application embodiment may include high-throughput sequencing, which can sequence hundreds of thousands to millions of DNA molecules at once, improving sequencing efficiency and reducing sequencing costs.

[0059] Transcriptome sequencing data of the library was analyzed, and expression profile and differential gene analysis were performed on the transcriptome data of HepG2 cells in the treatment group and the control group to obtain the genomic expression profile of human liver cells after the predicted drug exposure.

[0060] In one possible implementation, the raw sequencing data can be screened to improve data quality by removing sequencing adapters and low-quality data, resulting in high-quality Clean Reads. These Clean Reads are then subjected to genome-wide alignment analysis to determine the specific location of the sample data within the reference genome.

[0061] S102: Based on the expression profile, calculate the sum of the absolute values ​​of the fold differences of genes contained in the current gene pathway, and use this sum as the toxicity score of the current gene pathway.

[0062] In the embodiments of this application, the sum of the absolute values ​​of the fold differences of genes contained in the current gene pathway can be calculated based on the expression profile, and used as the toxicity score of the current gene pathway.

[0063] Specifically, 16 pathways related to hepatotoxicity were selected in the embodiments of this application, as shown in Table 1 below.

[0064]

[0065] Table 1

[0066] You can select any one of the 16 gene pathways mentioned above as the current gene pathway. You can perform differential gene fold analysis on the current gene pathway data of human liver cells after exposure to the drug to be predicted and the transcriptome data of the control group cells to obtain the differential fold value logFC (FoldChange) of the genes contained in the current gene pathway.

[0067] The absolute values ​​of the differential fold values ​​of genes contained in the current gene pathway can be added together to obtain the toxicity score of the current gene pathway.

[0068] In one possible implementation, the p-value can be calculated using the null hypothesis of the toxicity score of the current gene pathway. The p-value is defined as the probability that the toxicity score of the current gene pathway is greater than the toxicity score of a randomly selected set of gene pathways (e.g., it can be calculated 10,000 times randomly). The significance of each toxic pathway is estimated by the p-value. When p is less than 0.05, it indicates that the toxicity score of the current gene pathway is high, and the current gene pathway can be used for subsequent hepatotoxicity prediction.

[0069] S103: Calculate the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times.

[0070] S104: When the Z value is greater than 1.5, the drug to be predicted is predicted to have hepatotoxicity.

[0071] In this embodiment of the application, the Z-value of the current gene pathway can be calculated based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times; when the Z-value is greater than 1.5, the drug to be predicted is predicted to have hepatotoxicity.

[0072] Specifically, ;

[0073] Where Z is the Z-value, X is the toxicity score of the current gene pathway, μ0 is the mean, and σ is the standard deviation. The preset gene pathways can be randomly selected from all gene pathways, and are not limited to the 16 gene pathways selected above. Multiple calculations can be performed, for example, 10,000 calculations of the toxicity scores of the screened gene pathways.

[0074] When the Z-value of the current gene pathway is >1.5, it indicates that the drug to be predicted is significantly associated with this toxic pathway and has potential hepatotoxicity.

[0075] For example, when the drug to be predicted is aristolactam II, aristolactam II can be dissolved in DMSO to prepare a 100mM stock solution, and hepG2 cells can be treated with final concentrations of 1000μM and 100μM.

[0076] See Table 2 for the Z-scores used to evaluate hepatotoxicity after treatment of HepG2 cells with 1000 μM and 100 μM aristolochic acid lactam II.

[0077]

[0078] Table 2

[0079] Using a Z-score > 1.5 as a threshold, the potential hepatotoxic pathways of aristolactam II were analyzed. If any pathway with a Z-score > 1.5 was detected within the set of 22 pathways, the substance was predicted to have potential hepatotoxicity. The detected toxic pathways were displayed using a heatmap, such as... Figure 2 As shown, at 1000 μM, the chemical carcinogenesis and regulation of hemopoiesis pathways were detected with aristolochic acid II, while at 100 μM, chemical carcinogenesis was detected.

[0080] Aristolochic acid is classified as a Group 1 carcinogen for humans by the WHO (World Health Organization) / IARC (International Agency for Research on Cancer). Researchers in Taiwan published a paper in the authoritative medical journal *Science Translational Medicine*, confirming that aristolochic acid can also cause liver cancer. Aristolochic acid is metabolized in the body to aristolactam, which binds to DNA to form stable, difficult-to-degrade adducts, leading to DNA damage and inducing carcinogenesis. The hepatotoxicity prediction method of this invention also accurately predicted the chemical carcinogenic effects of aristolactam II, demonstrating its potential hepatotoxicity.

[0081] For example, when the drug to be predicted is blister beetle, cocklebur, or atractylodes, these three traditional Chinese medicines can be prepared by water extraction. The water extract is then freeze-dried to prepare a lyophilized powder. The lyophilized powders of the three traditional Chinese medicines are dissolved in PBS to prepare a concentration of 50 mg / mL. HepG2 cells are then treated with this concentration of 200 μg / mL.

[0082] See Table 3 for the Z-values ​​used to evaluate the hepatotoxicity of HepG2 cells after treatment with aqueous extracts of blister beetle, cocklebur, or atractylodes.

[0083]

[0084] Table 3

[0085] Using a Z-score > 1.5 as a threshold, the potential hepatotoxic pathways of blister beetles, cocklebur, and atractylodes were analyzed. If any pathway with a Z-score > 1.5 was detected within the set of 16 pathways, the substance was predicted to have potential hepatotoxicity. The detected toxic pathways were displayed using a heatmap, such as... Figure 3 As shown in the figure, blister beetles showed Z-scores > 1.5 in two pathways: Liver Fibrosis and Lymphocytic Infiltration, indicating potentially strong hepatotoxicity. Xanthium sibiricum also showed a Z-score > 1.5 in the Liver Fibrosis pathway, suggesting potentially strong hepatotoxicity, mainly associated with liver fibrosis. These toxicity results are largely consistent with clinical toxicity findings. Atractylodes lancea extract, however, did not show a Z-score > 1.5 in any pathway at 200 μg / mL, indicating that its hepatotoxicity is relatively low at this concentration. (Details follow...) Figure 3 As shown.

[0086] This application provides a method for predicting drug hepatotoxicity. The method includes: sequencing human liver cells exposed to the drug to be predicted to obtain a genomic expression profile of the liver cells; calculating the sum of the absolute values ​​of fold changes in genes within the current gene pathway based on the expression profile, as the toxicity score of the current gene pathway; calculating the Z-score of the current gene pathway based on the toxicity score of the current gene pathway, the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times; and predicting hepatotoxicity of the drug when the Z-score is greater than 1.5. Therefore, this application can accurately, timely, effectively, and sensitively predict drug hepatotoxicity without requiring a large number of live animals as experimental subjects, offering advantages in terms of safety, environmental protection, and animal welfare. It also has advantages such as a short experimental cycle and low cost.

[0087] This method overcomes the limitations of traditional compound hepatotoxicity prediction methods, enabling early prediction of delayed hepatotoxicity through toxic pathways or gene sets. This provides a highly efficient, accurate, and rapid method for predicting the hepatotoxicity of compounds or single drugs during the toxicity and safety assessment process in new drug development and food safety. This method can also be used for the evaluation and prediction of hepatotoxicity in the processing or combination of toxic traditional Chinese medicines to reduce toxicity and enhance efficacy.

[0088] Exemplary device

[0089] See Figure 4 The diagram shown is a schematic of a drug hepatotoxicity prediction device provided in an embodiment of this application, comprising:

[0090] Sequencing unit 201 is used to sequence human liver cells exposed to the drug to be predicted, and to obtain the genomic expression profile of the human liver cells exposed to the drug to be predicted.

[0091] The toxicity score calculation unit 202 is used to calculate the sum of the absolute values ​​of the fold differences of genes contained in the current gene pathway based on the expression profile, and use this sum as the toxicity score of the current gene pathway.

[0092] Z-value calculation unit 203 is used to calculate the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times.

[0093] The prediction unit 204 is used to predict that the drug to be predicted has hepatotoxicity when the Z value is greater than 1.5.

[0094] In one possible implementation, the Z-value calculation unit is specifically used to calculate the Z-value of the current gene pathway:

[0095] ;

[0096] Where Z is the Z-value, X is the toxicity score of the current gene pathway, μ0 is the mean, and σ is the standard deviation.

[0097] In one possible implementation, the current gene pathway includes: cell cycle gene pathway, replication gene pathway, chemical oncogenic gene pathway, apoptosis gene pathway, lysosomal gene pathway, necroptosis gene pathway, p53 signaling gene pathway, cellular senescence gene pathway, bile duct hyperplasia gene pathway, liver fibrosis gene pathway, liver steatosis gene pathway, cell division regulation gene pathway, hematopoietic regulation gene pathway, glycerophospholipid synthesis gene pathway, cell necrosis gene pathway, or leukocyte infiltration gene pathway.

[0098] In one possible implementation, the sequencing includes high-throughput sequencing.

[0099] This application provides a drug hepatotoxicity prediction device. The method of applying this device includes: sequencing human liver cells exposed to the drug to be predicted to obtain a genomic expression profile of the liver cells; calculating the sum of the absolute values ​​of the fold changes of genes in the current gene pathway based on the expression profile, as the toxicity score of the current gene pathway; calculating the Z-score of the current gene pathway based on the toxicity score of the current gene pathway, the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times; and predicting hepatotoxicity of the drug when the Z-score is greater than 1.5. Therefore, this application can accurately, timely, effectively, and sensitively predict drug hepatotoxicity without requiring a large number of live animals as an experimental basis, offering advantages in safety, environmental protection, and animal welfare. It also has advantages such as a short experimental cycle and low cost.

[0100] Based on the above embodiments, this application provides a drug hepatotoxicity prediction system, including:

[0101] Memory, used to store computer programs;

[0102] A processor is used to implement the steps of the drug hepatotoxicity prediction method described above when executing the computer program.

[0103] Based on the above embodiments, this application also provides a computer-readable medium storing a computer program, which, when processed and executed, implements the steps of the drug hepatotoxicity prediction method described above.

[0104] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0105] The aforementioned computer-readable medium may be included in the aforementioned system, or it may exist independently and not assembled into the system.

[0106] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts.

[0107] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0108] The above description is merely a preferred embodiment of this application. Although this application has disclosed preferred embodiments above, it is not intended to limit this application. Any person skilled in the art can make many possible variations and modifications to the technical solutions of this application using the methods and techniques disclosed above, or modify them into equivalent embodiments with equivalent changes, without departing from the scope of the technical solutions of this application. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of this application without departing from the content of the technical solutions of this application shall still fall within the protection scope of the technical solutions of this application.

Claims

1. A method for predicting drug hepatotoxicity, characterized in that, include: Human liver cells exposed to the drug to be predicted were sequenced to obtain the genomic expression profile of the human liver cells after exposure to the drug to be predicted. Based on the expression profile, the sum of the absolute values ​​of the differential fold values ​​of genes contained in the current gene pathway is calculated as the toxicity score of the current gene pathway. Specifically, human liver cells are used as control cells with the solvent of the drug to be predicted added. Differential fold analysis is performed on the current gene pathway data of the human liver cells after contact with the drug to be predicted and the transcriptome data of the control cells to obtain the differential fold values ​​(logFC) of genes contained in the current gene pathway. Then, the absolute values ​​of the differential fold values ​​of genes contained in the current gene pathway are summed as the toxicity score of the current gene pathway. The current gene pathway includes: cell cycle gene pathways, DNA replication gene pathways, chemical oncogenic gene pathways, apoptosis gene pathways, lysosomal gene pathways, necroptosis gene pathways, p53 signaling gene pathways, cellular senescence gene pathways, bile duct hyperplasia gene pathways, liver fibrosis gene pathways, liver steatosis gene pathways, cell division regulation gene pathways, hematopoietic regulation gene pathways, glycerophospholipid synthesis gene pathways, cell necrosis gene pathways, or leukocyte infiltration gene pathways. The Z-value of the current gene pathway is calculated based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times. When the Z value is greater than 1.5, the drug to be predicted is predicted to have hepatotoxicity. After calculating the sum of the absolute values ​​of the fold change values ​​of genes contained in the current gene pathway based on the expression profile, as the toxicity score of the current gene pathway, the method further includes: A null hypothesis p-value is calculated based on the toxicity score of the current gene pathway; wherein the p-value is the probability that the toxicity score of the current gene pathway is greater than the toxicity score of a randomly selected set of gene pathways. The step of calculating the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times includes: When the p value is less than 0.05, the Z value of the current gene pathway is calculated based on the toxicity score of the current gene pathway, the average value and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times. The step of calculating the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times includes: ； Where Z is the Z-value, X is the toxicity score of the current gene pathway, μ0 is the mean, and σ is the standard deviation.

2. The method according to claim 1, characterized in that, The sequencing includes high-throughput sequencing.

3. A drug hepatotoxicity prediction device, characterized in that, include: A sequencing unit is used to sequence human liver cells exposed to the drug to be predicted, and to obtain the genomic expression profile of the human liver cells exposed to the drug to be predicted. The toxicity score calculation unit is used to calculate the sum of the absolute values ​​of the differential fold values ​​of genes contained in the current gene pathway based on the expression profile, as the toxicity score of the current gene pathway. Specifically, human liver cells are added to the solvent of the drug to be predicted as control cells. Differential fold analysis is performed on the current gene pathway data of the human liver cells after contact with the drug to be predicted and the transcriptome data of the control cells to obtain the differential fold value (logFC) of genes contained in the current gene pathway. Then, the absolute values ​​of the differential fold values ​​of genes contained in the current gene pathway are summed to obtain the toxicity score of the current gene pathway. The current gene pathway includes: cell cycle gene pathways, DNA replication gene pathways, chemical oncogenic gene pathways, apoptosis gene pathways, lysosomal gene pathways, necroptosis gene pathways, p53 signaling gene pathways, cellular senescence gene pathways, bile duct hyperplasia gene pathways, liver fibrosis gene pathways, liver steatosis gene pathways, cell division regulation gene pathways, hematopoietic regulation gene pathways, glycerophospholipid synthesis gene pathways, cell necrosis gene pathways, or leukocyte infiltration gene pathways. The Z-value calculation unit is used to calculate the Z-value of the current gene pathway based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times. A prediction unit is used to predict that the drug to be predicted has hepatotoxicity when the Z value is greater than 1.

5. The toxicity score calculation unit is also used to calculate a p-value based on the toxicity score of the current gene pathway using a null hypothesis; wherein the p-value is the probability that the toxicity score of the current gene pathway is greater than the toxicity score of a randomly selected set of gene pathways. The Z-value calculation unit is specifically used to calculate the Z-value of the current gene pathway when the p-value is less than 0.05, based on the toxicity score of the current gene pathway and the average and standard deviation of the toxicity scores of multiple preset gene pathways calculated multiple times. The Z-value calculation unit is specifically used to calculate the Z-value of the current gene pathway: ； Where Z is the Z-value, X is the toxicity score of the current gene pathway, μ0 is the mean, and σ is the standard deviation.

4. The apparatus according to claim 3, characterized in that, The sequencing includes high-throughput sequencing.

5. A drug hepatotoxicity prediction system, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the drug hepatotoxicity prediction method as described in claim 1 or 2 when executing the computer program.

6. A computer-readable medium, characterized in that, The computer-readable medium stores a computer program that, when processed and executed, implements the steps of the drug hepatotoxicity prediction method as described in claim 1 or 2.

Images