Search

CN-122012716-A - Tumor early detection method and system based on DNA features

CN122012716ACN 122012716 ACN122012716 ACN 122012716ACN-122012716-A

Abstract

The invention discloses a tumor early detection method and system based on DNA features, which relates to the technical field of molecular diagnosis, and synchronously collects plasma samples at a plurality of time points by setting a standardized time sequence sample collection node, so as to ensure that the dynamic process of DNA steady state change can be captured, in the aspect of sample processing, cfDNA is extracted by adopting a magnetic bead method automatic extraction technology, strict quality control standard is set, the quality of a detection sample is ensured, the method comprises the steps of constructing an integrated mixed reaction system without cross interference, synchronously capturing oxidative damage marks and methylation modification in cfDNA (cfDNA) by using a specific probe, realizing synchronous detection of a two-dimensional molecular marker, acquiring accurate two-dimensional time sequence signal data by a two-channel fluorescence synchronous detection and internal reference correction technology, constructing a normal physiological dynamic reference model, and calculating a time sequence characteristic deviation value based on the model to realize accurate judgment of early tumor risk.

Inventors

  • LIU LIXIN
  • LI XIA
  • YUAN XIAOJIANG
  • SU QIN

Assignees

  • 深圳市本源生物科技有限公司

Dates

Publication Date
20260512
Application Date
20260226

Claims (10)

  1. 1. The early tumor detection method based on the DNA features is characterized by comprising the following specific steps: s1, collecting peripheral venous blood of a subject according to a preset physiological cycle time sequence node, separating to obtain a plasma sample, and extracting free cfDNA in the plasma; S2, performing synchronous hybridization capture on cfDNA by adopting an oxidative damage specific probe and a methylation specific probe, wherein the oxidative damage specific probe targets and binds with an 8-hydroxydeoxyguanosine modification site in the cfDNA, and the methylation specific probe targets and binds with a CpG island methylation site of a tumor associated gene; S3, quantitatively detecting the captured cfDNA, and respectively obtaining oxidation damage abundance and methylation level signals of the cfDNA under each time sequence node to form a two-dimensional time sequence characteristic data set; S4, based on the two-dimensional time sequence detection data in the physiological cycle of the healthy crowd, establishing a DNA oxidative damage-methylation dynamic change reference model in a normal physiological state, and determining a time sequence characteristic deviation threshold; S5, comparing the two-dimensional time sequence feature data set of the detected person with a reference model, calculating a time sequence feature deviation value, and if the deviation value exceeds a preset threshold value, judging that the detected person has abnormal DNA steady state, and prompting early tumor occurrence risk.
  2. 2. The method for early detection of tumors based on DNA features of claim 1, wherein in the step S1, four continuous time sequence detection nodes T0, T1, T2 and T3 are set on the basis of the physiological rhythm of a detected subject, the interval duration of each node is unified to 48 hours, 5 mL/node of peripheral venous blood of the detected subject is collected by adopting an EDTA-K2 anticoagulation vacuum blood collection tube, a sample is placed in a condition of 4 ℃ and 3000g for centrifugation for 15 minutes within 1 hour after blood collection to complete plasma separation, upper cell-free plasma is sucked and transferred to an enzyme-free centrifuge tube, and the upper cell-free plasma is frozen and stored in an environment of-80 ℃ for standby.
  3. 3. The early tumor detection method based on DNA features of claim 1, wherein in S1, the specific steps of extracting free cfDNA in plasma are that a magnetic bead method is adopted to automatically extract plasma cfDNA, 200 mu L of thawed plasma sample is taken, 400 mu L of lysis buffer containing 20mM Tris-HCl, 8M guanidine salt and 1% beta-mercaptoethanol is added, after 30S of vortex oscillation, room temperature is used for lysis for 10min, then 20 mu L of carboxyl modified magnetic beads are added and fully and uniformly oscillated, room temperature is used for incubation for 15min to combine cfDNA with the magnetic beads, supernatant after magnetic separation is added, washing buffer containing 70% ethanol and 10mM Tris-HCl is used for magnetic separation washing, 2 times of washing is repeated to remove impurities, finally 50 mu L of 10mM Tris-HCl non-enzyme elution buffer containing 50 mu L of pH8.0 is added, cfDNA is eluted after 10min of 65 ℃, quality control is carried out by using a Qubit4.0 fluorescent quantitative instrument, agarose gel electrophoresis and Nanodrop detector after 10min, the concentration of the cfDNA is more than or equal to 1 mu L, the main ratio is equal to or more than 200.0, and the main ratio is equal to or more than 200.0 bp/200.280.
  4. 4. The method for early detection of tumor based on DNA features according to claim 1, wherein in S2, a 5' -end-labeled FAM fluorophore and an aptamer probe targeting 8-hydroxydeoxyguanosine are selected as oxidative damage specific probes, a recombinant methylation binding domain protein fusion probe with an N-end-labeled HEX fluorophore is selected as methylation specific probes, a reaction system with a total volume of 50 μl is prepared, the reaction system comprises a 10 μ LcfDNA template, 2 μl10 μM oxidative damage probes, 2 μl10 μM methylation probes, 25 μl2×hybridization buffer containing 50mMPBS, 1MNaCl, 0.1% Tween-20 and 11 μl anhydrous enzyme, the reaction system is uniformly mixed and denatured at 95 ℃ for 5min, quenched in an ice bath for 2min, hybridized at constant temperature for 90min at 42 ℃,10 μl streptavidin magnetic beads are added after hybridization, incubation is performed at room temperature for 10min, free probes are magnetically separated and washed 3 times with a washing buffer, and a DNA complex combined with the dual-dimensional probes is retained.
  5. 5. The method for early detection of tumors based on DNA features of claim 1, wherein in the step S3, a dual-channel fluorescence synchronous detection and internal reference correction technology is adopted, oxidation damage abundance and methylation level signals are synchronously acquired through a FAM channel and a HEX channel of a real-time fluorescence quantitative PCR instrument, a beta-actin housekeeping gene without oxidation damage and without methylation is introduced as an internal reference, systematic error correction is carried out on a target detection signal by taking a fluorescence signal of the beta-actin housekeeping gene as a reference, corrected FAM fluorescence values and HEX fluorescence values of four time sequence nodes from T0 to T3 are respectively recorded, and a two-dimensional time sequence data matrix comprising oxidation damage and methylation levels is constructed.
  6. 6. The method for early detection of tumors based on DNA features of claim 1, wherein in S4, healthy volunteers are selected, a training data set is constructed by acquiring two-dimensional time sequence data according to a whole process from time sequence sample acquisition to time sequence signal quantitative detection, a dynamic time warping algorithm is adopted to fit the time sequence data of healthy people, an average dynamic change track of DNA oxidative damage-methylation under a normal physiological state is generated, and a time sequence feature deviation threshold value is determined to be 1.25 through ROC curve analysis and is used as a judgment standard for distinguishing normal physiological fluctuation from pathological abnormality.
  7. 7. The method for early detection of tumors based on DNA features of claim 6, wherein in S4, the specific steps of fitting the time series data of healthy people by adopting a dynamic time warping algorithm are that a two-dimensional time series data set of healthy volunteers is selected, oxidation injury abundance and methylation level data corresponding to each time series node of each volunteer are preprocessed to obtain standardized individual two-dimensional time series tracks, the time series track mean value of partial healthy volunteers is selected as an initial reference track by adopting the dynamic time warping algorithm, the individual time series tracks of other healthy volunteers are respectively and dynamically aligned with the initial reference track, the corresponding relation of each time series node is adjusted by calculating the accumulated distance between every two tracks, the average value of oxidation injury abundance and methylation level under each time series node is calculated based on all the aligned individual time series tracks, the track alignment and reference track updating steps are repeated until the reference tracks tend to be stable, namely, the track fluctuation after two adjacent updating reaches the stable standard, the operation is stopped, and the finally obtained reference track is the dynamic average change of the DNA oxidation injury-methylation track under the preset normal physiological state.
  8. 8. The method for early detection of tumor based on DNA features of claim 1, wherein in S5, the two-dimensional time series feature dataset of the subject is compared with a reference model, and a time series feature deviation value is calculated, and the calculation formula is: , wherein, The time sequence characteristic deviation value represents the overall deviation degree of the two-dimensional time sequence track of the detected person and the reference track of the healthy crowd; Is the set total number of timing detection nodes, The time sequence node traversal sequence number is used for sequentially referring to each independent sampling time sequence node; To the subject at the first Under each time sequence node, DNA oxidation damage abundance quantitative values corrected by internal references, Is in the average dynamic reference track of healthy people, the first DNA oxidation damage abundance reference value corresponding to each time sequence node, Is subject at the first Under each time sequence node, DNA methylation level quantitative values corrected by internal references, Is in the average dynamic reference track of healthy people, the first The DNA methylation level reference value corresponding to each time sequence node is obtained by fitting a healthy crowd data set through a dynamic time warping algorithm.
  9. 9. The method for early detection of tumor based on DNA features according to claim 8, wherein in S5, Is the timing characteristic deviation value of the signal, Less than or equal to 1.25, and 1.25 </SUB > is judged as normal DNA steady state without early tumor risk Less than or equal to 1.8, judging that the DNA steady state is slightly abnormal, prompting the risk of precancerous lesions and suggesting a 3-month later review, 1.8, Severe abnormalities in DNA homeostasis suggested a high risk for early stage tumors and suggested further imaging and pathological examination.
  10. 10. A DNA signature based tumor early detection system adapted for use in a DNA signature based tumor early detection method according to any one of claims 1-8, said system comprising: The sample pretreatment module is used for separating peripheral blood and plasma, extracting and purifying cfDNA and outputting a standardized cfDNA sample; the double-dimensional probe capturing module integrates an oxidation damage specific probe and a methylation specific probe reaction system to realize synchronous specific capturing of cfDNA double-dimensional modification sites; The time sequence data acquisition module is connected with the fluorescence detection unit, acquires oxidation injury abundance and methylation level signals of cfDNA according to a preset time sequence node, and generates a time sequence characteristic data set; the dynamic model analysis module is internally provided with a two-dimensional time sequence reference model library of healthy people, is provided with a time sequence deviation calculation unit and is used for completing comparison analysis of data of a detected person and a reference model; The result output module is used for outputting DNA steady state abnormality judgment results, time sequence characteristic deviation curves and early tumor risk classification reports.

Description

Tumor early detection method and system based on DNA features Technical Field The invention relates to the technical field of molecular diagnosis, in particular to a tumor early detection method and system based on DNA features. Background In recent years, along with the continuous progress of molecular biology technology, molecular diagnosis shows great potential in the field of early detection of tumors, wherein plasma free DNA is taken as circulating nucleic acid released into blood by tumor cells, contains rich tumor-related information, genomic DNA can suffer oxidative damage in early stage of tumorigenesis, and the generation of oxidative damage markers such as 8-hydroxydeoxyguanosine can be caused, and can directly reflect the unbalanced state of genomic oxidative stress, and meanwhile, methylation modification of tumor-related gene CpG islands is also a key epigenetic event in the tumorigenesis process, can regulate the expression of oncogenes and tumor suppressor genes, and plays an important role in the initiation and progression stages of tumors. However, current tumor fluid biopsy techniques have significant limitations. Most detection schemes focus on molecular markers with single dimension, either detect oxidative damage or methylation modification, and cannot realize synchronous capture and joint analysis of the two-dimensional molecular markers, and the single-dimension detection mode leads to incomplete characterization of DNA steady-state abnormality and easy omission of important early tumor information. Meanwhile, the static detection mode of a single time point is generally adopted in the prior art, a normal dynamic reference standard in a physiological period is not established, molecular level in a human body can be influenced by various physiological factors to fluctuate, and the single time point detection cannot distinguish physiological molecular fluctuation from pathological evolution characteristics, so that the identification sensitivity of early critical lesions of tumors is insufficient, false negative and false positive judgment results are easy to appear. Disclosure of Invention The invention aims to make up the defects of the prior art, and provides a tumor early detection method and system based on DNA features, which can synchronously collect plasma samples at a plurality of time points by setting a standardized time sequence sample collection node, ensure that a dynamic process of DNA steady state change can be captured, extract cfDNA (cfDNA) by adopting a magnetic bead method automatic extraction technology in the aspect of sample processing, set strict quality control standard, ensure the quality of detection samples, construct an integrated mixed reaction system without cross interference, synchronously capture oxidative damage marks and methylation modification in the cfDNA by using a specific probe, realize synchronous detection of a two-dimensional molecular marker, acquire accurate two-dimensional time sequence signal data by a two-channel fluorescence synchronous detection and internal reference correction technology, construct a normal physiological dynamic reference model, and realize accurate judgment of tumor early risk by calculating a time sequence feature deviation value based on the model. In order to solve the technical problems, the invention provides the following technical scheme that, on one hand, the method for detecting the early tumor based on the DNA features comprises the following specific steps: s1, collecting peripheral venous blood of a subject according to a preset physiological cycle time sequence node, separating to obtain a plasma sample, and extracting free cfDNA in the plasma; S2, performing synchronous hybridization capture on cfDNA by adopting an oxidative damage specific probe and a methylation specific probe, wherein the oxidative damage specific probe targets and binds with an 8-hydroxydeoxyguanosine modification site in the cfDNA, and the methylation specific probe targets and binds with a CpG island methylation site of a tumor associated gene; S3, quantitatively detecting the captured cfDNA, and respectively obtaining oxidation damage abundance and methylation level signals of the cfDNA under each time sequence node to form a two-dimensional time sequence characteristic data set; S4, based on the two-dimensional time sequence detection data in the physiological cycle of the healthy crowd, establishing a DNA oxidative damage-methylation dynamic change reference model in a normal physiological state, and determining a time sequence characteristic deviation threshold; S5, comparing the two-dimensional time sequence feature data set of the detected person with a reference model, calculating a time sequence feature deviation value, and if the deviation value exceeds a preset threshold value, judging that the detected person has abnormal DNA steady state, and prompting early tumor occurrence risk. Further, in the step S1, four contin