Platelet RNA marker combination for detecting non-small cell lung cancer and application of platelet RNA marker combination

Abstract

The invention discloses a platelet RNA marker combination for detecting non-small cell lung cancer and application of the platelet RNA marker combination, and belongs to the technical field of biological medicine. According to the invention, platelet transcriptome data of a public database NSCLC patient and a healthy control person are compared and analyzed, bioinformatics screening, ROC curve analysis and single center queue verification are carried out, and values of three platelet RNA markers including peroxidase PRDX4, monoamine oxidase MAOB and glutathione transferase GSTM5 in non-small cell lung cancer diagnosis are finally determined. A non-small cell lung cancer diagnosis model based on the platelet RNA marker is successfully constructed, and the vacancy of the platelet RNA marker in the construction of the non-small cell lung cancer diagnosis model is filled.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technologies, and more specifically, to a platelet RNA biomarker combination for detecting non-small cell lung cancer and its application. Background Art

[0002] Lung cancer is the leading cause of cancer death globally, and non-small cell lung cancer (NSCLC) accounts for more than 80% of all lung cancers. Although significant progress has been made in the treatment of lung cancer in recent years, the 5-year survival rate of this disease remains low, mainly due to the lack of effective early diagnosis methods. The National Lung Screening Trial (NLST) in the United States and the Netherlands-Belgium Lung Screening Trial (NELSON) low-dose CT screening trials have shown that early detection can reduce mortality. Currently, common methods for tumor diagnosis include pathological tissue biopsy, low-dose computed tomography (LDCT), positron emission tomography-computed tomography (PET-CT), and liquid biopsy. However, pathological tissue biopsy can cause trauma, pain, and discomfort to patients and may pose a certain risk of infection; percutaneous aspiration has a certain probability of causing needle-track tumor seeding. Although LDCT can reduce the mortality of lung cancer patients, it has disadvantages such as poor imaging, high false positive rate, and radiation exposure. PET-CT examination has high sensitivity and specificity in the diagnosis, staging, and treatment evaluation of lung cancer, but its equipment is scarce and expensive. Liquid biopsy has the characteristics of minimally invasive, fast, and high sensitivity, and is accepted by most patients. With the rapid development of LB technology, the detection of tumor markers has shifted from static detection to dynamic monitoring, and the research content and direction have gradually become rich. Common tumor markers for lung cancer include carcinoembryonic antigen, NSE, carcinoembryonic antigen, cytokeratin 19 fragment, SCC, etc., but there is currently no specific tumor marker for non-small cell lung cancer.

[0003] In summary, the existing non-small cell lung cancer diagnosis methods have problems such as high invasiveness, high cost, insufficient sensitivity, lack of specific tumor markers, and inability to detect the disease in the early stage. Platelets, as carriers of circulating biomarkers, play a key role in the regulation of the tumor microenvironment by their RNA expression profiles, but there is rarely the development of an NSCLC diagnosis model based on platelet RNA biomarkers. Moreover, traditional molecular markers can usually only effectively detect patients with advanced lung cancer, and the positive rate for the diagnosis of stage I lung cancer is less than 10%. The 5-year survival rate of patients with early non-small cell lung cancer can reach more than 80%, while the 5-year survival rate of patients with advanced non-small cell lung cancer is less than 20%. Improving the detection rate of early non-small cell lung cancer is of great significance for improving the prognosis of non-small cell lung cancer.

[0004] Therefore, how to provide a platelet RNA biomarker for detecting non-small cell lung cancer and its application to diagnose non-small cell lung cancer in the early stage is a technical problem that those skilled in the art urgently need to solve. Summary of the Invention

[0005] In view of this, the present invention provides a platelet RNA biomarker combination for detecting non-small cell lung cancer and its application, and successfully constructs a non-small cell lung cancer diagnosis model. The detection accuracy of early non-small cell lung cancer by this model is 77.8%, which is of great significance for improving the prognosis of non-small cell lung cancer.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A platelet RNA biomarker combination for detecting non-small cell lung cancer, including any one, two or three combinations of glutathione transferase GSTM5, peroxidase PRDX4 and monoamine oxidase MAOB.

[0007] Another object of the present invention is to provide: the use of the above non-small cell lung cancer platelet RNA biomarker combination, and the use includes any one of the following: 1) Application in the preparation of products for diagnosing non-small cell lung cancer; 2) Application in the preparation of drugs for treating and / or preventing non-small cell lung cancer; 3) Application in establishing a model for remotely monitoring and / or diagnosing the development or prognosis of non-small cell lung cancer.

[0008] Another object of the present invention is to provide: a non-small cell lung cancer detection product, including the above platelet RNA biomarker combination.

[0009] As a preferred technical solution, the product includes but is not limited to a qRT-PCR detection kit, a digital PCR detection chip, a nanopore sequencing rapid diagnosis device, and a computer model.

[0010] Another object of the present invention is to provide: a drug for treating and / or preventing non-small cell lung cancer, the drug including the above platelet RNA biomarker combination and a pharmaceutically acceptable carrier; The carrier includes excipients, disintegrants, buffers, and stabilizers.

[0011] Another object of the present invention is to provide: a computer model for predicting and / or monitoring and / or diagnosing non-small cell lung cancer, the model receiving expression data of multiple RNA biomarkers as input and outputting the sensitivity probability or risk score of an individual; The model is established using machine learning algorithms; As a preferred technical solution, the algorithm is selected from random forest, logistic regression, LASSO regression, COX regression, artificial neural network, decision tree, support vector machine, naive Bayes, K-nearest neighbor algorithm, gradient boosting tree, Adaboost algorithm, XGBoost algorithm, LightGBM algorithm, CatBoost algorithm, multi-layer perceptron, convolutional neural network, recurrent neural network, long short-term memory network, gated recurrent unit network, etc.; Wherein the RNA marker is the above-mentioned marker combination.

[0012] Another object of the present invention is to provide: a method for constructing a non-small cell lung cancer diagnosis model based on platelet RNA markers, comprising the following steps: (1) Screening platelet RNAs with co-differential expression in platelets and tissues of non-small cell lung cancer patients from a bioinformatics database; (2) Collecting whole blood samples of untreated non-small cell lung cancer patients, performing platelet separation, extraction and purity quality control; (3) Extracting and purifying platelet RNA to remove genomic DNA; (4) Reverse transcribing platelet RNA to obtain cDNA, and performing amplification to obtain the original data of CT values, and calculating the expression level of platelet RNA for constructing the model; (5) Performing original data processing of platelet RNA expression levels, cleaning the original data required for constructing the model, and performing differential expression analysis of platelet RNA; (6) Evaluating the diagnostic value of platelet RNA in non-small cell lung cancer, selecting the features used for constructing the model and optimizing the model parameters, and constructing a non-small cell lung cancer diagnosis model based on platelet RNA markers. It can be seen from the above technical solutions that compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a platelet RNA marker for detecting non-small cell lung cancer and its application. Aiming at the technical bottlenecks such as insufficient specificity in the early diagnosis of non-small cell lung cancer (NSCLC) and poor compliance of invasive detection methods in the prior art, the present invention first reveals the diagnostic value of three platelet RNA markers, PRDX4, MAOB and GSTM5, in non-small cell lung cancer through a technical route combining analysis of public data transcriptome sequencing and algorithms. Based on clinical verification, the above markers have excellent AUC, high sensitivity and good specificity, and can be widely used in the development of non-small cell lung cancer auxiliary diagnostic reagents. Compared with traditional tissue biopsy, this method only needs to collect peripheral blood samples to achieve non-invasive detection. Brief Description of the Drawings

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to the provided drawings.

[0014] Appendix Figure 1 It is: Differentially expressed genes in platelets of non-small cell lung cancer patients provided by the embodiments of the present invention. Among them, a: Volcano plot of differentially expressed platelet RNA of NSCLC patients based on the public database GSE68086 provided by the embodiments of the present invention; b: MA plot of differentially expressed platelet RNA of NSCLC patients based on the public database GSE68086 provided by the embodiments of the present invention; c: Venn diagram of the GEPIA2 dataset and the GSE68086 dataset of up-regulated genes in non-small cell lung cancer patients provided by the embodiments of the present invention; d: Venn diagram of the GEPIA2 dataset and the GSE68086 dataset of down-regulated genes in non-small cell lung cancer patients provided by the embodiments of the present invention.

[0015] Appendix Figure 2 It is: Platelet purity detection results. Among them, a: Results of flow cytometry for identifying platelet purity; b: Result analysis diagram of flow cytometry.

[0016] Appendix Figure 3 It is: Schematic diagram of differential expression of PRDX4, MAOB, and GSTM5 in platelets of non-small cell lung cancer and healthy people in TEPNSCLC provided by the present invention; among them, a: represents PRDX4; b: represents MAOB; c: represents GSTM5.

[0017] Appendix Figure 4 It is: Flowchart of the method for constructing a diagnostic model for non-small cell lung cancer using platelet GSTM5, PRDX4, and MAOB genes provided by the embodiments of the present invention.

[0018] Appendix Figure 5 It is: Schematic diagram of the diagnostic efficacy of using PRDX4, MAOB, and GSTM5 to assist in diagnosing non-small cell lung cancer patients by ROC curve analysis.

[0019] Appendix Figure 6 It is: Schematic diagram of the diagnostic efficacy of using PRDX4, MAOB, and GSTM5 to assist in diagnosing non-small cell lung cancer patients by ROC curve analysis. Among them, a-c: are schematic diagrams of the diagnostic efficacy of pairwise combination of PRDX4, MAOB, and GSTM5 to assist in diagnosing non-small cell lung cancer patients by ROC curve analysis; d: Schematic diagram of the diagnostic efficacy of combined use of PRDX4, MAOB, and GSTM5 to assist in diagnosing non-small cell lung cancer patients by ROC curve analysis. Detailed implementation manners

[0020] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0021] Example 1, Preliminary screening of platelet RNA markers related to non-small cell lung cancer 1) Screen the TEPRNA with co-differential expression in platelets and tissues of non-small cell lung cancer patients from the biobank, specifically as follows:

[0022] Download the NSCLC dataset GSE68086 from the GEO database and preliminarily screen the differential genes between platelets of non-small cell lung cancer patients and platelets of healthy normal people in the dataset; Download the differential genes of tumor tissues and normal tissues of lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) from the GEPIA2 database.

[0023] As shown in the appendix Figure 1 Combined with the GEO and GEPIA2 databases, screen the TEP RNA of co-differentially expressed genes in platelets and tissues of stage IA lung adenocarcinoma and lung squamous cell carcinoma patients: First, refer to the differential expression mRNA data (GEO: GSE68086) between healthy donor platelets and non-small cell lung cancer tumor-educated platelets published in 2015 and draw volcano plots and MA plots. At the same time, the up-regulated and down-regulated genes of lung cancer were obtained from GEPIA 2. Subsequently, genes with p value <0.01, Log FC > 1.5 or < -1.5 in the two datasets were further analyzed through the Venn dataset, and their diagnostic potential and efficacy were analyzed. 23 markers such as platelet PRDX4, MAOB, and GSTM5 were screened as differentially expressed genes and preliminarily determined for the diagnosis of non-small cell lung cancer (TEPNSCLC). The Ensembl gene IDs in the Ensembl database are as follows: ENSG00000123131, ENSG00000069535, ENSG00000134201.

[0024] 2) Collect whole blood samples of untreated non-small cell lung cancer patients for platelet separation, extraction, and purity quality control Collect more than 2 mL of EDTA-K2 anticoagulated whole blood from untreated non-small cell lung cancer patients. Centrifuge the EDTA-K2 anticoagulated whole blood at 120 g for 10 min at room temperature. Transfer the supernatant to another 1.5 ml centrifuge tube, label it, centrifuge at 120 g for 10 min again, and then transfer the supernatant to a 500 μl centrifuge tube to detect the platelet purity. The detection method is as follows: Perform platelet counting with a blood cell counter. If the number of nucleated cells in every 10 million platelets is <5, the purity is considered good.

[0025] Then transfer the supernatant to a 1.5 ml centrifuge tube, label it, centrifuge at 360 g for 20 min, discard the supernatant, and retain the precipitate to obtain platelets with higher purity.

[0026] To verify the platelet purity, use flow cytometry to detect platelet markers CD41 and CD42a. The method is as follows: Step 1: Prepare a human washed platelet suspension and adjust its concentration to 1×10 7 cells / ml. Then take 50 μl of the platelet suspension and add it to a 0.5 ml centrifuge tube. Add 20 μl of CD41-PE and 20 μl of CD42a FITC for staining, and incubate in the dark for 30 min to obtain a stained platelet suspension. Step 2: Add 400 μl of Tyrode Buffer to the stained platelet suspension, then transfer it to a flow tube, and measure the percentage of CD41 and CD42a antibody positivity by flow cytometry. The results are as shown in the appendix Figure 2 As shown, the expression rates of CD41 and CD42a in platelets are both above 90%, indicating that the platelets extracted by differential centrifugation have good purity.

[0027] 3) Extract and purify platelet RNA to remove genomic DNA

[0028] ① Use the Qiagen miRNeasy Micro Kit to extract RNA from platelets and store the RNA at -80 °C: Add 700 μl of QiAzol lysis buffer to the platelet sample extracted in 1) above, pipette and vortex until there are no white lumps, let it stand at 20 - 25 °C for 6 min, then add 140 μl of chloroform, tightly cap the tube, vigorously vortex for 15 s, let it stand at room temperature for 3 min, and then centrifuge at 4 °C, 12000 g for 5 min.

[0029] Transfer 350 μl of the upper aqueous phase obtained by centrifugation to a 1.5 ml centrifuge tube (avoid any interphase), add 525 μl of absolute ethanol, mix well by inverting up and down, and then transfer it to an RNeasy MinElute spin column, close the lid, and centrifuge at 20 - 25 °C, ≥8000 g for 15 s.

[0030] Add the flowing-through liquid back into the spin column to improve RNA recovery rate, and repeat the operation. Discard the flowing-through liquid, add the remaining liquid, repeat the operation, and discard the flowing-through liquid.

[0031] Add 700 µl of Buffer RWT eluent, close the lid and incubate at room temperature for 1 min, centrifuge at 10,000 g at room temperature for 15 s, and discard the flowing-through liquid; add 500 µl of RPE, close the lid and incubate at room temperature for 1 min, centrifuge at 10,000 g for 15 s, and discard the flowing-through liquid. Add 500 µl of 80% ethanol, close the lid and incubate at room temperature for 1 min, centrifuge at 10,000 g for 2 min, and discard the flowing-through liquid and the collection tube.

[0032] Place the RNeasy MinElue spin column into a new 2-ml collection tube. Open the lid and centrifuge at full speed (MAX) for 5 min to dry the membrane, and discard the waste liquid and the collection tube.

[0033] Place the spin column into a new 1.5-ml collection tube, add 10 µl of RNase-free water to the center of the spin membrane. Cover the lid (place the lid of the collection tube clockwise), elute RNA at full speed (MAX) for 1 min, measure the RNA concentration, and record the RNA concentration and A260 / A280 (a purity of 1.9 - 2.1 is considered good).

[0034] ② Purify platelet RNA and remove genomic DNA. The specific operations are as follows: Use the Vazyme HiScript Ⅳ RT SuperMix for qPCR (+gDNA wiper) kit, gently pipette and mix well, place it in a gene amplifier, and incubate at 42 °C for 2 min to purify RNA and remove genomic DNA (gDNA). The gDNA removal reaction system is shown in Table 1 below Table 1 gDNA removal reaction system

[0035] 4) Reverse transcribe platelet RNA to obtain cDNA, and perform amplification to obtain the original data of CT values, and calculate the expression level of the platelet RNA for constructing the model ① Reverse transcribe platelet RNA to cDNA, specifically as follows: Reverse transcribe RNA into cDNA using the Vazyme HiScript Ⅳ RT SuperMix for qPCR (+gDNA wiper) kit. Prepare the reverse transcription reaction system as shown in Table 2 below and perform reverse transcription. The reaction conditions are as follows: 50 °C for 15 min; 85 °C for 5 s. After the reaction, store at -20 °C for later use (store at 20 °C for half a year and at -80 °C for long term).

[0036] Table 2 Reverse transcription reaction system

[0037] ② cDNA amplification; Obtain the raw data of CT values: Amplify cDNA using the ChamQ Blue Universal SYBR qPCR Master Mix qPCR kit. The following operations and reagent preparations are all carried out on ice. The specific qPCR reaction system is shown in Table 3, the primer sequences for amplification are shown in Table 4, and the specific reaction conditions are shown in Table 5.

[0038] Table 3 qPCR reaction system

[0039] Table 4 Upstream and downstream primer sequences

[0040] Table 5 qPCR reaction conditions

[0041] ③ Calculate the expression level: Use 2- ΔCt For the quantitative data analysis of the two target genes, ΔCt = Ct value of the target gene - Ct of the internal reference gene. The Ct value of the target gene is the Ct value of the target genes PRDX4, MAOB, GSTM5 detected by real-time fluorescence quantitative technology. Calculate their respective relative expression levels using GAPDH as the internal reference.

[0042] 5) Perform the processing of the raw data of platelet RNA expression levels, clean the raw data required for model construction, and perform differential expression analysis of platelet RNA; Use the non-parametric Mann-Whitney U test to analyze the relative expression levels of platelet RNA. The results are shown in Appendix Figure 3 and Table 6.

[0043] Table 6 Comparison of differential gene expression levels of platelets between the case group and the control group

[0044] Result analysis: The results are shown in AppendixFigure 3 As shown in Table 6, the expression levels of PRDX4 and MAOB in the platelets of non-small cell lung cancer patients were higher than those in the healthy control group, and the differences were statistically significant. The expression level of GSTM5 in the platelets of non-small cell lung cancer patients was lower than that in the healthy control group, and the difference was statistically significant. Therefore, it was finally determined that PRDX4, MAOB, and GSTM5 could be used as platelet RNA markers for non-small cell lung cancer patients.

[0045] Example 2: Construction of a non-small cell lung cancer model based on platelet RNA markers Select the features used to construct the model and optimize the model parameters to construct a non-small cell lung cancer diagnosis model based on platelet RNA markers (attached Figure 4 ), specifically as follows: ① ROC diagnostic efficacy evaluation ROC analysis was performed on the diagnostic values of the target genes PRDX4, MAOB, and GSTM5 in platelets for non-small cell lung cancer. The experimental results are shown in the attached Figure 5 , attached Figure 6 and Table 7.

[0046] Table 7 Evaluation of the efficacy of platelet differential genes in diagnosing non-small cell lung cancer

[0047] Result analysis: As can be seen from the results, GSTM5 in platelets had better diagnostic efficacy than PRDX4 and MAOB. However, when the three mRNAs were combined for diagnosis and a diagnostic model (TEPNSCLC) was established, the diagnostic efficacy was significantly higher than that of the individual TEP RNA, and the sensitivity reached more than 80%, indicating that the combined expression levels of PRDX4, MAOB, and GSTM5 in platelets could better diagnose NSCLC.

[0048] ② Univariate Logistic analysis of clinical indicators and platelet differential genes in non-small cell lung cancer patients and healthy individuals was as follows: Table 8 Univariate Logistic analysis of clinical indicators and platelet differential genes in non-small cell lung cancer patients and healthy individuals

[0049] Result analysis: As can be seen from the results, PRDX4, MAOB, and GSTM5 were statistically significant (P < 0.05); while gender, age, smoking status, and platelet DEFA3 were not statistically significant.

[0050] ③ Multivariate Logistic analysis of platelet differential genes in non-small cell lung cancer patients and healthy individuals was as follows: Table 9 Multivariate Logistic analysis of platelet differential genes in non-small cell lung cancer patients and healthy individuals

[0051] The meaningful results of the univariate analysis were included in the multivariate analysis, and platelet PRDX4, MAOB, and GSTM5 were included to construct a multivariate Logistic regression equation (Method: input). Result analysis: The effect of platelet MAOB on NSCLC was statistically significant (OR = 5.282, 95%CI 1.605 - 17.381, P < 0.006); the effect of platelet GSTM5 on NSCLC was statistically significant (OR = 0.327, 95%CI 0.161 - 0.664, P < 0.002).

[0052] ④Construct a diagnostic model According to the results of the Logistic regression analysis, the clinical prediction probability model for NSCLC was obtained: P = e (-1.680+0.561X1+1.664X2-1.117X3) / [1 + e (-1.680+0.561X1+1.664X2-1.117X3) , (P: probability of NSCLC occurrence, X1: relative expression level of platelet PRDX4, X2: relative expression level of platelet MAOB, X3: relative expression level of platelet GSTM5).

[0053] Table 10 Prediction results of the clinical prediction probability model for NSCLC

[0054] The correct percentage of the model was 77.8%. The Hosmer - Lemeshow goodness - of - fit test was performed on the prediction model, χ² = 6.541, P = 0.587 > 0.05. The model had good fitting validity and had a certain predictive value.

[0055] In this specification, each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0056] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A platelet RNA biomarker combination for detecting non-small cell lung cancer, characterized in that, It includes any one, two or three combinations of glutathione transferase GSTM5, peroxidase PRDX4 and monoamine oxidase MAOB.

2. Use of the non-small cell lung cancer platelet RNA marker combination according to claim 1, characterized in that, The uses include any one of the following: 1) Application in the preparation of products for diagnosing non-small cell lung cancer; 2) Application in the preparation of drugs for treating and / or preventing non-small cell lung cancer; 3) Application in the establishment of a model for remotely monitoring and / or diagnosing the development or prognosis of non-small cell lung cancer.

3. A non-small cell lung cancer detection product, characterized in that, It contains the platelet RNA marker combination described in claim 1.

4. The product according to claim 3, characterized in that, The products include, but are not limited to, qRT-PCR detection kits, digital PCR detection chips, nanopore sequencing rapid diagnostic devices, and computer models.

5. A drug for treating and / or preventing non-small cell lung cancer, characterized in that, The drugs include the platelet RNA marker combination described in claim 1 and a pharmaceutically acceptable carrier; The carrier includes excipients, disintegrants, buffers, and stabilizers.

6. A computer model for predicting and / or monitoring and / or diagnosing non-small cell lung cancer, characterized in that, The model receives the expression data of multiple RNA markers as input and outputs the sensitivity probability or risk score of an individual; The model is established using machine learning algorithms; Preferably, the algorithms are selected from random forest, logistic regression, LASSO regression, COX regression, artificial neural network, decision tree, support vector machine, naive Bayes, K-nearest neighbor algorithm, gradient boosting tree, Adaboost algorithm, XGBoost algorithm, LightGBM algorithm, CatBoost algorithm, multi-layer perceptron, convolutional neural network, recurrent neural network, long short-term memory network, gated recurrent unit network, etc.; Wherein the RNA marker is the marker combination described in claim 1.

7. A method for constructing a diagnostic model for non-small cell lung cancer based on platelet RNA markers, characterized in that, It includes the following steps: (1) Screen the platelet RNA that is commonly differentially expressed in the platelets and tissues of non-small cell lung cancer patients from a biobank; (2) Collect whole blood samples from untreated non-small cell lung cancer patients, perform platelet separation, extraction, and purity quality control; (3) Extract and purify the platelet RNA to remove genomic DNA; (4) Reverse transcribe the platelet RNA to obtain cDNA, and perform amplification to obtain the original data of CT values, and calculate the expression level of the platelet RNA for constructing the model; (5) Perform the original data processing of the platelet RNA expression level, clean the original data required for constructing the model, and perform differential expression analysis of the platelet RNA; (6) Evaluate the diagnostic value of the platelet RNA in non-small cell lung cancer, select the features used for constructing the model and optimize the model parameters, and construct a non-small cell lung cancer diagnostic model based on the platelet RNA marker.

Images