Comprehensive risk assessment method based on microbial whole genome sequencing

Abstract

The invention discloses a comprehensive risk assessment method based on microorganism whole genome sequencing, and belongs to the technical field of medicine production and microorganism safety control. According to the method, high-precision species identification is performed through kmerfinder, drug-resistant genes are identified in combination with RGI, and multi-dimensional information such as clinical detection frequency, disease relevance, spore / biofilm forming ability and drug-resistant gene public health risk is integrated to construct a quantitative scoring system; and according to the normalized total score, dividing the microbial pollution risk into S, A, B, C and D levels, performing correction in combination with a list of unacceptable microorganisms in pharmacopeia, and finally outputting a standardized risk level report. The method realizes scientific, objective and accurate grading of microbial pollution risks in a pharmaceutical environment, and is suitable for microbial monitoring and decision support under a GMP system.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of microbial genomics and biosafety risk assessment, and specifically relates to a comprehensive method based on microbial whole-genome sequencing data, integrating species identification, drug-resistant gene analysis and multi-dimensional risk assessment, which is particularly suitable for microbial contamination risk assessment and management in the fields of pharmaceutical production environment, clinical medicine and public health. BACKGROUND

[0002] With the in-depth implementation of the "whole life cycle" quality control concept of pharmaceuticals, enterprises need to accurately identify and trace the origin of the microorganisms detected in drug raw materials, excipients, intermediate products, final products, pharmaceutical water, environment, etc. The bacterial DNA characteristic sequence identification method in Chapter 1021 of Chinese Pharmacopoeia 2025 Edition points out that the standard system of molecular biology detection technology is constructed based on first-generation nucleic acid sequencing technology (Sanger method), and for some closely related bacterial species that cannot be accurately identified by using 16S rRNA gene characteristic nucleic acid sequence, molecular biology identification can be combined with whole-genome sequence to meet the increasingly stringent requirements of pharmaceutical microbial quality control.

[0003] Microbial whole-genome sequencing is a process of using high-throughput sequencing technology to determine the entire genome sequence of a microbial individual and obtaining genetic information. It is a new generation of nucleic acid sequencing technology that can obtain the most abundant and comprehensive genetic information of a strain, and has shown great advantages in the accurate identification, traceability analysis and risk assessment of closely related species of microorganisms in the field of pharmaceuticals. At present, the second-generation and third-generation high-throughput nucleic acid sequencing technology has gradually matured and is applied in various fields such as testing and detection, basic science and clinical diagnosis. However, the existing technology mainly focuses on the identification and traceability of microorganisms, lacks a systematic, comprehensive and quantifiable microbial risk assessment method, and cannot directly convert genomic sequencing data into risk levels with practical guiding significance, making it difficult to make scientific decisions in pharmaceutical quality control and clinical infection prevention and control.

[0004] The existing microbial risk assessment methods have the following shortcomings: 1. Single evaluation dimension, only focusing on one aspect of drug resistance or pathogenicity, lacking a comprehensive evaluation system; 2. The evaluation standard is highly subjective, lacking quantitative indicators, resulting in poor comparability of the evaluation results; 3. Public database resources are not fully integrated, and massive clinical and environmental isolate data cannot be utilized; 4. The influence of the special biological characteristics of microorganisms (such as spore-forming ability and biofilm-forming ability) on risk is not considered; 5. The evaluation results are not highly related to the actual clinical hazards, and it is difficult to guide the actual risk prevention and control measures.

[0005] Therefore, it is urgent to develop a comprehensive risk assessment method based on whole genome sequencing data of microorganisms, to realize the scientific transformation from genome data to risk level, and to provide scientific basis for drug quality control, clinical infection prevention and control, and public health decision-making. SUMMARY

[0006] The purpose of the present application is to provide a comprehensive risk assessment method based on whole genome sequencing of microorganisms, to establish a scientific, objective and quantifiable microorganism risk assessment system by integrating species identification, drug resistance gene analysis, clinical disease association, environmental distribution characteristics and microbial biological characteristics, and to realize accurate assessment and hierarchical management of microbial contamination risk.

[0007] The scheme for realizing the technical purpose of the present application is as follows: A comprehensive risk assessment method based on whole genome sequencing of microorganisms, comprising the following steps: (1) Using kmerfinder algorithm to identify species of whole genome sequencing data of microorganisms, to obtain species name, taxonomic ID and taxonomic hierarchy information; (2) Using RGI tool to identify drug resistance genes based on CARD database, to obtain drug resistance gene type, number and ARO classification information; (3) Based on the results of steps (1) and (2), a multi-dimensional comprehensive scoring system is constructed: a) Isolate number and distribution evaluation: Obtain the number of clinical isolates (clinical) and the number of environmental isolates (environmental) through NCBI Pathogen database, calculate the total number of isolates total = clinical + environmental; When total > 100,000, 3 points are given, when 30,000 < total ≤ 100,000, 2 points are given, when 10,000 < total ≤ 30,000, 1 point is given, and when total ≤ 10,000, 0 point is given; Clinical isolate proportion score = (clinical / total) × 5, maximum value is 5 points; b) Clinical disease association evaluation: Query whether the microorganism is associated with a specific clinical disease through PathoPhenoDB database and NCBI Pathogen database, if there is a disease association record with evidence level of "confirmed", add 2 points; c) Microbial biological characteristic evaluation: using BLASTp to align the genomic sequence with sporulation-related genes (spo0A, sigF) and biofilm formation-related genes (bap, icaA), if there is at least one sporulation-related gene, spore_score = 2, otherwise 0; if there is at least one biofilm formation-related gene, biofilm_score = 2, otherwise 0; Characteristic_Score = spore_score + biofilm_score; d) Drug resistance gene risk evaluation: based on the ARO system in the CARD database, each drug resistance gene is assigned a Q1, Q2, Q3, Q4 or RI = 0 risk level, where Q1 = 10 points, Q2 = 7 points, Q3 = 4 points, Q4 = 1 point, RI = 0 = 0 point; Calculate the weighted average score = (Q1 number x 10 + Q2 number x 7 + Q3 number x 4 + Q4 number x 1 + RI0 number x 0) / total number of drug resistance genes; ; (4) Calculate the original risk total score Raw_Score = isolate number and distribution score + clinical disease association score + Characteristic_Score + drug resistance gene risk evaluation score; (5) If the microorganism is listed in the list of unacceptable microorganisms in the current version of the pharmacopoeia of the target country, then Specialized_Score = Raw_Score x 1.2; otherwise Specialized_Score = Raw_Score; (6) Normalization process, which normalizes the weighted average score into a fixed interval, making it easy to risk grade: Final_Score = (Specialized_Score - min_Score) x 20 / (max_Score - min_Score) (7) Determine the risk level according to Final_Score, the range of Final_Score is set to 0-20: S level ≥ 16.0; A level 12.0 ≤ Final_Score < 16.0; B level 8.0 ≤ Final_Score < 12.0; C level 4.0 ≤ Final_Score < 8.0; D level Final_Score 4.0; (8) Output a standardized report containing the species name, drug resistance gene list, Raw_Score, Final_Score and risk level.

[0008] In step (3)a), the proportion score of clinical isolates is 0 when total = 0.

[0009] In step (3)b), the explicit clinical disease association includes pneumonia, sepsis or wound infection.

[0010] In step (3)c), the sporulation-related genes also include sspB and cotA, and the biofilm formation-related genes also include pel and psl.

[0011] In step (3)d), RI = 0 indicates that the drug-resistant gene has no public health risk.

[0012] The specific species in the list of unacceptable microorganisms includes Pseudomonas aeruginosa, Staphylococcus aureus or Candida albicans.

[0013] The method is applied to the risk assessment of microbial contamination in pharmaceutical production environment, sterile preparation workshop or pharmaceutical water system.

[0014] The beneficial effects of the present application are: 1. Multi-dimensional comprehensive evaluation: The present application integrates the evaluation indexes of four dimensions of clinical disease association, environmental distribution characteristics, drug-resistant gene risk and microbial biological characteristics, overcomes the defect of single evaluation dimension in the prior art, and realizes the comprehensive evaluation of microbial risk.

[0015] 2. Objective and quantitative evaluation: By establishing a scientific scoring system and calculation formula, the complex microbial characteristics are converted into quantifiable risk scores, avoiding the problem of strong subjectivity in traditional evaluation methods, and improving the objectivity and comparability of the evaluation results.

[0016] 3. Data-driven decision making: Make full use of the massive data resources of public databases such as PathoPhenoDB and NCBI Pathogen, so that the evaluation results are based on real-world big data, enhancing the scientificity and reliability of the evaluation.

[0017] 4. Precise risk classification: Through a five-level risk classification system (S, A, B, C, D), the precise classification of microbial risk is realized, providing clear action guidelines for drug quality control, clinical infection prevention and control, and public health decision making.

[0018] 5. Simple and efficient operation: The method is based on automatic analysis of whole genome sequencing data, which can be seamlessly integrated with existing microbial sequencing processes, and the evaluation process is highly automated, greatly improving the efficiency of risk assessment.

[0019] 6. Wide range of applications: Suitable for drug production environment monitoring, clinical infection pathogen identification, public health epidemic warning and other scenarios, with wide practical value. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the method of the present invention.

[0021] Figure 2 This is a schematic diagram of the microbial risk assessment scoring system, which shows in detail the composition of the four core assessment modules and their interrelationships.

[0022] Figure 3 This is a risk score distribution diagram of different microorganisms in the examples. Detailed Implementation

[0023] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be noted that the following embodiments are only used to explain the present invention, but the scope of protection of the present invention is not limited to these embodiments.

[0024] Example 1: Implementation of a comprehensive risk assessment method based on microbial whole-genome sequencing Figure 1 This is a flowchart illustrating the method of the present invention, showing the complete process from input of whole genome sequencing data to output of the final risk assessment report. Figure 2 This is a schematic diagram of the microbial risk assessment scoring system, which shows in detail the composition of the four core assessment modules and their interrelationships.

[0025] 1. Sample preparation and whole-genome sequencing: The target microorganism was isolated from the material delivery port of the drug; genomic DNA of the microorganism was extracted, and whole-genome sequencing was performed using the Illumina NovaSeq 6000 sequencing platform to obtain PE150 paired-end sequencing data; sequencing data quality control: FastQC was used for quality assessment, and Trimmomatic was used to remove low-quality reads and adapter sequences to obtain quality-controlled data.

[0026] 2. Microbial species identification: High-quality sequences were assembled using Unicycler (v0.4.8) software, and the quality-controlled sequencing data were identified using the k-merfinder v2.2 tool. The k-mer length was set to the default, and the alignment database was the RefSeq complete genome database. Species name, taxonomic ID (taxid), and taxonomic hierarchy information were obtained. Example results: Acinetobacter baumannii (taxid: 470), Staphylococcus aureus (taxid: 1280), etc.

[0027] 3. Antimicrobial resistance gene identification: The genome annotation tool rgi (v5.2.1) was used to analyze antimicrobial resistance genes in whole genome data; the database used was CARD v3.2.6, with an identity threshold of 90% and a coverage threshold of 80%; antimicrobial resistance genes in microbial genomes were identified and annotated to obtain the types, quantities and distribution characteristics of antimicrobial resistance genes; Example results: AB10 strain was found to contain resistance genes for multidrug (12 types) and aminoglycoside antibiotic (9 types).

[0028] 4. Antimicrobial resistance gene risk assessment: Obtain the risk level (Q1-Q4 and RI=0) of each common antimicrobial resistance gene from the article database; taking AB10 strain as an example: 38 antimicrobial resistance genes were detected, of which 5 were at level Q1, 15 at level Q2, 12 at level Q3, and 6 at level Q4; calculate the weighted average score = (5×10+15×7+12×4+6×1) / 38 = (50+105+48+6) / 38 = 209 / 38 = 3.5 points; based on the weighted average score of 5.5 points, the risk level of the antimicrobial resistance gene is determined to be level B; this score of 5.5 points is used as the score for this module.

[0029] 5. Clinical disease association assessment: The association information between microorganisms and clinical diseases was queried through the PathoPhenoDB database; AB10 was mapped to "Acinetobacter baumannii pneumonia" and "Acutepneumonia" in the database, and was not NA or null; 2 points were added for the presence of the disease.

[0030] 6. Environmental Distribution and Clinical Isolation Characteristics Assessment: Clinical and environmental isolate data of microorganisms were obtained from the NCBI Pathogen database; the number of AB10 clinical isolates was 36,916, and the number of environmental isolates was 1,430; the total number of isolates = 36,916 + 1,430 = 38,346 > 10,000, receiving 3 points; the proportion of clinical isolates = 36,916 / 38,346 = 0.963, multiplied by 3, yielding 2.889 points; the total score for the assessment of isolate quantity and distribution = 3 + 2.889 = 5.889 points. Example result: Ec6 strains are distributed in multiple countries including the USA (166,494) and the United Kingdom (62,738), with a Distribution_Score of 9.2.

[0031] 7. Assessment of microbial biological characteristics: Analysis of the biological characteristics possessed by the species; Taking strain AB10 as an example: spore_forming="No", biofilm_forming="Yes"; scores are based on the presence and expression level of relevant genes: spore_score = 2 if spore formation ability is present; biofilm_score = 2 if biofilm formation ability is present; otherwise, 0 points are awarded, Characteristic_Score = spore_score + biofilm_score; strain AB10 does not have spore formation ability but has biofilm formation ability, so Characteristic_Score is 2.

[0032] 8. Comprehensive risk level determination and risk correction: Specialized_Score = Raw Score × Correction coefficient; The correction factor is based on the acceptance of certain microorganisms according to the pharmacopoeias of various countries. If it is an "unacceptable microorganism", then the correction factor = original score × 1.2.

[0033] The final score is calculated using normalization: Final_Score = (Specialized_Score - min_Score) × 20 / (max_Score - min_Score) Risk level is determined based on Final Score: S-level: Final_Score ≥ 16.0 Grade A: 12.0 ≤ Final_Score < 16.0 Grade B: 8.0 ≤ Final_Score < 12.0 Grade C: 4.0 ≤ Final_Score < 8.0 Grade D: Final_Score < 4.0 Generate a comprehensive assessment report that includes species information, drug resistance characteristics, risk score, and risk level.

[0034] Figure 3 The risk score distribution map of different microorganisms in the examples intuitively demonstrates the ability of this method to distinguish the risks of different microorganisms.

[0035] Example result: PA5 strain is Pseudomonas aeruginosa Pseudomonas aeruginosa is explicitly marked as an unacceptable microorganism in the pharmacopoeias of some countries. The final score should be multiplied by a correction factor of 1.2, resulting in a final score of 14.0175 * 1.2 = 16.821, which is rated as S.

[0036] Through the above steps, this invention integrates multi-dimensional information such as species identification, drug resistance gene analysis, clinical disease correlation, environmental distribution characteristics, and microbial biological characteristics by analyzing the whole genome. It efficiently and quickly completes a comprehensive risk assessment of environmental microorganisms in pharmaceutical companies, realizing the scientific transformation from genomic data to risk levels, and providing a scientific basis for drug quality control, clinical infection prevention and control, and public health decision-making.

[0037] The embodiments described above can be further combined or replaced, and these embodiments are merely descriptions of preferred embodiments of the present invention, not limitations on the concept and scope of the present invention. Various changes and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the inventive concept are all within the protection scope of the present invention. The protection scope of the present invention is given by the appended claims and any equivalents.

Claims

1. A comprehensive risk assessment method based on whole genome sequencing of microorganisms, characterized in that, Comprising the following steps: (1) Species identification of microbial whole genome sequencing data using kmerfinder algorithm, obtaining species name, taxonomic ID and taxonomic rank information; (2) Using RGI tool to identify drug-resistant genes based on CARD database to obtain drug-resistant gene type, number and ARO classification information; (3) Based on the results of steps (1) and (2), a multi-dimensional comprehensive scoring system is constructed: a) Isolate number and distribution evaluation: Obtain the number of clinical isolates (clinical) and environmental isolates (environmental) through the NCBI Pathogen database, calculate the total number of isolates total = clinical + environmental; When total > 100,000, 3 points are given, when 30,000 < total ≤ 100,000, 2 points are given, when 10,000 < total ≤ 30,000, 1 point is given, and when total ≤ 10,000, 0 point is given; Clinical isolate ratio score = (clinical / total) × 5, maximum value is 5 points; b) Clinical disease association evaluation: Query whether the microorganism is associated with a specific clinical disease through the PathoPhenoDB database and the NCBI Pathogen database, if there is a record of disease association with the evidence level of "confirmed", add 2 points; c) Microbial biological characteristics evaluation: Use BLASTp to align the genomic sequence with the spore formation related genes spo0A, sigF and biofilm formation related genes bap, icaA, if there is at least one spore formation related gene, then spore_score = 2, otherwise 0; If there is at least one biofilm formation related gene, then biofilm_score = 2, otherwise 0; Characteristic_Score = spore_score + biofilm_score; d) Drug-resistant gene risk evaluation: Based on the ARO system in the CARD database, each drug-resistant gene is assigned a Q1, Q2, Q3, Q4 or RI=0 risk level, where Q1=10 points, Q2=7 points, Q3=4 points, Q4=1 point, RI=0=0 point; Calculate the weighted average score = (Q1 number × 10 + Q2 number × 7 + Q3 number × 4 + Q4 number × 1 + RI0 number × 0) / total number of drug-resistant genes; ； (4) Calculate the original risk total score Raw_Score = isolate number and distribution score + clinical disease association score + Characteristic_Score + drug-resistant gene risk evaluation score; (5) If the microorganism is listed in the unacceptable microorganism list of the current edition of the target country pharmacopoeia, then Specialized_Score = Raw_Score x 1.2; otherwise Specialized_Score = Raw_Score; (6) Normalization process, which is to normalize the weighted average score into a fixed interval, to facilitate risk grading: Final_Score = (Specialized_Score - min_Score) x 20 / (max_Score - min_Score) (7) Determine the risk level according to Final_Score, the range of Final_Score is set to 0-20: S level ≥ 16.0; A level 12.0 ≤ Final_Score < 16.0; B level 8.0 ≤ Final_Score < 12.0; C level 4.0 ≤ Final_Score < 8.0; D level Final_Score 4.0; (8) Output the standardized report containing the species name, drug resistance gene list, Raw_Score, Final_Score and risk level.

2. The method of claim 1, wherein, In step (3)a), when total = 0, the proportion score of clinical isolates is 0.

3. The method of claim 1, wherein, In step (3)b), the clear clinical disease association includes pneumonia, sepsis or wound infection.

4. The method of claim 1, wherein, In step (3)c), the spore formation related genes also include sspB, cotA; the biofilm formation related genes also include pel, psl.

5. The method of claim 1, wherein, In step (3)d), RI = 0 indicates that there is no public health risk of drug resistance genes.

6. The method of claim 1, wherein, The specific species in the unacceptable microorganism list includes Pseudomonas aeruginosa, Staphylococcus aureus or Candida albicans.

7. The method of claim 1, wherein, Applied to the risk assessment of microbial contamination in pharmaceutical production environment, sterile preparation workshop or pharmaceutical water system.

Images