A renal clear cell carcinoma diagnostic kit based on long non-coding RNA and its preparation method and application

Abstract

The present invention provides a renal clear cell carcinoma diagnostic kit based on long non-coding RNA and its application, which has the characteristics of high accuracy, high specificity and high sensitivity, can accurately diagnose renal clear cell carcinoma, and then achieve targeted guidance of individualized treatment, and has high clinical application value.

Description

Technical Field

[0001] The present invention belongs to the technical field of bioinformatics, and in particular relates to a renal clear cell carcinoma diagnostic kit based on long-chain non-coding RNA and an application thereof. Background Art

[0002] Kidney cancer, also known as renal cell carcinoma (RCC), is a malignant tumor originating from the renal parenchymal urinary tubule epithelial system. It is also the most common renal parenchymal malignant tumor, accounting for about 2% to 3% of adult malignant tumors. Renal cell carcinoma includes subtypes such as clear cell carcinoma, papillary renal cell carcinoma, chromophobe cell carcinoma, collecting duct carcinoma and undifferentiated renal cell carcinoma. Among them, kidney renal clear cell carcinoma (KIRC) is the most common type, accounting for 75% to 85% of all renal cell carcinoma patients. Current evidence shows that its pathogenesis is still inconclusive, and it may be related to genetics, smoking, obesity, hypertension and the use of antihypertensive drugs. Its typical manifestation is the "renal cancer triad", but it is currently rare in clinical practice. It is often discovered during imaging examinations due to hematuria, pain and lumps, so such patients are prone to miss the best treatment period.

[0003] Currently known non-coding RNAs mainly include microRNA, long non-coding RNA (lncRNA), circular RNA (circRNA), small nucleolar RNA (snoRNA), Piwi-interacting RNA (piRNA), etc. Long non-coding RNA is a type of non-coding RNA molecule with a length of more than 200 nt. It lacks an open reading frame (ORF), has no protein coding function, and has almost no protein coding ability. In recent years, studies have gradually found that long non-coding RNA is widely involved in almost all physiological and pathological activities of the human body, and participates in, regulates or mediates the occurrence and development of diseases. lncRNAs are very likely to become new biomarkers and therapeutic targets for cancer.

[0004] Despite significant progress in the diagnosis and treatment of renal cancer in recent years, renal clear cell carcinoma remains a challenging clinical problem due to its high heterogeneity and uncertain prognosis. No study has yet explored the correlation between renal clear cell carcinoma diagnosis and lncRNAs. Summary of the invention

[0005] In order to overcome the above problems, this study aims to develop a renal clear cell carcinoma diagnostic kit based on long noncoding RNA, in order to provide assistance in the clinical diagnosis of renal clear cell carcinoma.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solutions:

[0007] A diagnostic evaluation model for renal clear cell carcinoma based on long non-coding RNA, characterized in that a diagnostic score is calculated according to the model to diagnose renal clear cell carcinoma,

[0008] Diagnostic score = 1.8673322 + 0.5545214 * HIF1A-AS2 expression + 0.4002162 * C6orf223 expression + (-2.1904251) * AC019197.1 expression + (-0.7083016) * LINC01762 expression + 1.3368045 * AC097534.2 expression + 0.5440289 * DARS-AS1 expression;

[0009] Furthermore, the diagnostic score of 0 was used as the prognostic evaluation threshold, that is, a diagnostic score greater than or equal to 0 was renal clear cell carcinoma, and a diagnostic score less than 0 was non-renal clear cell carcinoma. The expression level was quantified as transcripts per million mapped reads per kilobase (TPM).

[0010] Furthermore, the present invention provides a method for constructing a renal clear cell carcinoma diagnostic evaluation model based on long non-coding RNA, comprising the following steps:

[0011] (1) Obtain tumor tissue and adjacent normal tissue samples from patients with renal clear cell carcinoma for whole transcriptome sequencing, and perform statistical analysis after homogenization using a standardization function;

[0012] (2) Screening for long noncoding RNAs with statistically significant differences in expression in tumor tissues relative to adjacent adjacent tissues;

[0013] (3) The cancer tissues and adjacent tissues were randomly divided into training sets and validation sets. The training set data was used to construct the diagnostic kit. The recursive feature elimination algorithm was used for 50 rounds of screening to finally determine the top 8 characteristic long non-coding RNAs in terms of frequency. The characteristic long non-coding RNAs with high frequency of occurrence were tested and selected in the LASSO regression analysis to construct the above-mentioned diagnostic evaluation model.

[0014] (4) The performance of the diagnostic evaluation model was verified and evaluated using validation set patient specimens and a public dataset (TCGA-KIRC dataset).

[0015] Furthermore, the patient's tumor tissue and adjacent tissue samples included in step (1) are from patients with renal clear cell carcinoma who underwent surgery in Wuhan Tongji Hospital, and the expression matrix of all long non-coding RNA in the cancer tissue and adjacent tissue is obtained by whole transcriptome sequencing.

[0016] Furthermore, in step (3), the characteristic long non-coding RNAs with high occurrence frequencies are AC097534.2, LINC0762, TTC21B.AS1, LINC00887, C6orf223, HIF1A.AS2, and AC019197.1.

[0017] Furthermore, the step (4) is specifically divided into the following steps:

[0018] Step (41) calculates the diagnostic score of each validation set patient specimen, classifies the patient specimens according to the diagnostic threshold of 0, draws the ROC curve and the expression heat map of the long non-coding RNA used to construct the model, and calculates the area under the curve;

[0019] Step (42) is verified again using a public dataset (TCGA-KIRC dataset), a ROC curve is drawn and a heat map of the expression of the long non-coding RNA used to construct the model is drawn, and the area under the curve is calculated.

[0020] The present invention further provides a product comprising the above-mentioned renal clear cell carcinoma diagnostic kit based on long non-coding RNA, wherein the product is a reagent, a chip or a kit.

[0021] The present invention further provides a method for preparing a product of a renal clear cell carcinoma diagnostic kit, wherein the product includes the above-mentioned renal clear cell carcinoma diagnostic kit based on long-chain non-coding RNA, and the product can be a reagent, a chip or a kit.

[0022] Beneficial effects:

[0023] The present invention provides a renal clear cell carcinoma diagnostic kit based on long non-coding RNA and its application, which has the characteristics of high accuracy, high specificity and high sensitivity, can accurately diagnose renal clear cell carcinoma, and then achieve targeted guidance of individualized treatment, and has high clinical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is the volcano map of differentially expressed genes in 100 cancer tissues and 50 adjacent normal tissues in the TJ-RCC cohort.

[0025] Figure 2 IncRNAs and their occurrence frequencies screened out after RF-RFF algorithm (10 folds) was repeated 50 times when constructing the prognostic model.

[0026] Figure 3 Plot of error frequency versus Log(λ) when building a prognostic model for LASSO regression.

[0027] Figure 4 Construct a distribution plot of coefficients in the prognostic model for LASSO regression.

[0028] Figure 5 is the corresponding coefficient of the long non-coding RNA used to construct the model.

[0029] Figure 6 Verify the situation for the TJ-RCC situation.

[0030] Figure 7 Validation status of patients in the TCGA-KIRC dataset. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0032] In some specific embodiments, a long non-coding RNA-based renal clear cell carcinoma diagnostic kit is provided, characterized in that a diagnostic score is calculated according to a model to diagnose renal clear cell carcinoma, and the kit comprises primers and PCR reagents required for detecting the expression levels of AC097534.2, LINC0762, TTC21B.AS1, LINC00887, C6orf223, HIF1A.AS2, and AC019197.1.

[0033] Diagnostic score = 1.8673322 + 0.5545214 * HIF1A-AS2 expression + 0.4002162 * C6orf223 expression + (-2.1904251) * AC019197.1 expression + (-0.7083016) * LINC01762 expression + 1.3368045 * AC097534.2 expression + 0.5440289 * DARS-AS1 expression;

[0034] The diagnostic score of 0 was used as the prognostic evaluation threshold, that is, a diagnostic score greater than or equal to 0 was considered renal clear cell carcinoma, and a diagnostic score less than 0 was considered non-renal clear cell carcinoma. The expression level was quantified as transcripts per kilobase of transcript per million mapped reads.

[0035] In another specific embodiment, the present invention provides a method for preparing a renal clear cell carcinoma diagnostic kit based on long non-coding RNA, which is characterized by obtaining the expression amount of long non-coding RNA, screening characteristic long non-coding RNA, and model construction, comprising the following steps:

[0036] (1) Obtain tumor tissue samples and adjacent normal tissues from patients with renal clear cell carcinoma for whole transcriptome sequencing, and perform statistical analysis after homogenization using a standardization function;

[0037] (2) Screening for long noncoding RNAs with statistically significant differences in expression in tumor tissues relative to adjacent adjacent tissues;

[0038] (3) The cancer tissues and adjacent tissues were randomly divided into training sets and validation sets. The training set data was used to construct the diagnostic kit. The repeated recursive feature elimination algorithm was used for 50 screenings to finally determine the top 8 characteristic long non-coding RNAs in terms of frequency. The characteristic long non-coding RNAs with high frequency were tested and selected in the LASSO regression analysis to construct the above-mentioned diagnostic evaluation model.

[0039] (4) The performance of the diagnostic evaluation model was verified and evaluated using the validation set patient specimens and the public dataset TCGA-KIRC dataset.

[0040] Among them, the patient's tumor tissue and adjacent tissue samples were from patients with renal clear cell carcinoma who underwent surgery in Wuhan Tongji Hospital. Whole transcriptome sequencing was used to obtain the expression matrix of all long non-coding RNAs in the cancer tissue and adjacent tissue.

[0041] As a preferred specific embodiment, the characteristic long-chain non-coding RNA with high occurrence frequency is specifically the top 8 characteristic long-chain non-coding RNA with the highest occurrence frequency, namely AC097534.2, LINC0762, TTC21B.AS1, LINC00887, C6orf223, HIF1A.AS2, and AC019197.1.

[0042] In another specific embodiment, the step (4) is specifically divided into the following steps:

[0043] Step (41) calculates the diagnostic score of each validation set patient specimen, classifies the patient specimens according to the diagnostic threshold of 0, draws the ROC curve and the expression heat map of the long non-coding RNA used to construct the model, and calculates the area under the curve;

[0044] Step (42) is verified again using a public dataset (TCGA-KIRC dataset), a ROC curve is drawn and a heat map of the expression of the long non-coding RNA used to construct the model is drawn, and the area under the curve is calculated.

[0045] In another specific embodiment, the present invention provides a product comprising the above-mentioned long non-coding RNA-based renal clear cell carcinoma diagnostic kit, wherein the product is a reagent, a chip or a kit.

[0046] In another specific embodiment, the present invention provides a method for preparing a product of a renal clear cell carcinoma diagnostic kit, wherein the product includes the above-mentioned renal clear cell carcinoma diagnostic kit based on long non-coding RNA, and the product can be a reagent, a chip or a kit.

[0047] In order to further clearly explain, illustrate and illustrate the present invention, a more specific description is given below in conjunction with the accompanying drawings.

[0048] Example 1

[0049] Figure 1 This is a volcano map of differentially expressed genes in 100 cancer tissues and 50 adjacent normal tissues in the TJ-RCC (Tongji Hospital-renal cell carcinoma patients) cohort. The tumor tissue and adjacent normal tissue samples of patients with renal clear cell carcinoma who underwent surgical treatment at Tongji Hospital in Wuhan were obtained for whole transcriptome sequencing, and the expression matrix of all long non-coding RNAs in 100 cancer tissues and 50 adjacent normal tissues was obtained. The normalization function preProcess(method=c("center","scale")) (where center represents the predicted variable value minus the mean; scale represents the predicted variable value divided by the standard deviation, so the default normalization method is (x-mu) / std) was used for normalization and statistical analysis was performed to screen long non-coding RNAs with statistically significant expressions in cancer tissues and adjacent normal tissues, such as Figure 1 The long noncoding RNAs with statistically significant expression differences between cancer tissues and adjacent paracancerous tissues are shown.

[0050] Example 2

[0051] Figure 2 When constructing the prognostic model, the long non-coding RNAs and their occurrence frequencies were screened out after 50 repetitions of the RF-RFF algorithm (10 fold). Specifically, the recursive feature elimination algorithm was used to repeat the screening 50 times, and finally the top 8 characteristic long non-coding RNAs with the highest occurrence frequency were determined for the construction of the diagnostic kit, such as Figure 2 .

[0052] Example 3

[0053] Cancer tissues and paracancerous tissues were randomly divided into training sets and validation sets. The training set data was used to construct a diagnostic kit, and the top 8 characteristic long non-coding RNAs with the highest frequency were tested and selected in LASSO regression analysis ( Figure 3 ) to construct a diagnostic assessment model and obtain

[0054] Diagnostic score = 1.8673322 + 0.5545214 * HIF1A-AS2 expression + 0.4002162 * C6orf223 expression + (-2.1904251) * AC019197.1 expression + (-0.7083016) * LINC01762 expression + 1.3368045 * AC097534.2 expression + 0.5440289 * DARS-AS1 expression ( Figure 4 );

[0055] The diagnostic threshold of 0 was determined based on the actual situation of patient specimens in the training set, that is, a diagnostic score greater than or equal to 0 was renal clear cell carcinoma, and a diagnostic score less than 0 was non-renal clear cell carcinoma. The expression level was quantified as transcripts per million (TPM) per kilobase of transcripts mapped per million reads.

[0056] Example 4

[0057] The diagnostic score of each validation set patient specimen was calculated according to the diagnostic score calculation formula, and the patient specimens were classified according to the diagnostic threshold of 0. The ROC curve and the expression heat map of the long non-coding RNA used in the model were drawn, and the area under the curve (AUC) was calculated. The AUC of the Wuhan Tongji Hospital validation set was 0.980 ( Figure 6 ), indicating that the diagnostic performance of the diagnostic kit is good. In order to further verify the performance of this diagnostic kit, the researchers used the public data set TCGA-KIRC for further verification, plotted the ROC curve and the expression heat map of the long non-coding RNA used to construct the model, and obtained the AUC of TCGA-KIRC = 0.936 ( Figure 7 ), once again proving that the diagnostic kit performs extremely well.

[0058] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any technician familiar with the present invention can make some changes or modify the technical contents suggested above into equivalent embodiments without departing from the technical scope of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the solution of the present invention.

Claims

1. A kit, characterized in that The kit contains reagents required to detect the expression levels of AC097534.2, LINC01762, TTC21B.AS1, LINC00887, C6orf223, HIF1A.AS2, AC019197.1, and DARS-AS1. The kit is used to diagnose and evaluate renal clear cell carcinoma based on the diagnostic score obtained by the test. Diagnostic score = 1.8673322 + 0.5545214 * HIF1A-AS2 expression + 0.4002162 * C6orf223 expression + (-2.1904251) * AC019197.1 expression + (-0.7083016) * LINC01762 expression + 1.3368045 * AC097534.2 expression + 0.5440289 * DARS-AS1 expression; the diagnostic score of 0 was used as the prognostic evaluation threshold, and the expression was quantified as transcripts per kilobase of transcription per million mapped reads.

Images