Methylation markers for detecting benign or malignant thyroid nodules and uses thereof

Abstract

The application belongs to the technical field of molecular biology and medicine, and discloses a methylation marker for detecting the benignity and malignancy of thyroid nodules and an application. 780 new methylation sites are identified, 273 tumor occurrence and development related genes are associated, and the Fc gamma R-mediated phagocytosis signal pathway is involved. The methylation marker for detecting the benignity and malignancy of thyroid nodules comprises multiple genes and multiple methylation sites. The methylation level of multiple genes and multiple sites is detected, the problem of low and unstable methylation signal of single gene or single site is overcome, and thus the sensitivity and specificity of detection are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of molecular biomedical technology, and particularly relates to a methylation marker for detecting the benignity or malignancy of thyroid nodules and use thereof. BACKGROUND

[0002] Thyroid cancer is the most common malignant tumor of the endocrine system. In recent years, the incidence of thyroid cancer has shown a rapid growth trend and has ranked among the top 10 malignant tumors. Thyroid cancer mainly includes four pathological types, i.e., papillary thyroid carcinoma (PTC), follicular thyroid carcinoma (FTC), anaplastic thyroid carcinoma (ATC) and medullary thyroid carcinoma (MTC), among which PTC is the most common pathological subtype, accounting for 75-80% of all thyroid cancers. At present, the clinical standard diagnosis and treatment process of thyroid cancer is to perform preoperative benign or malignant diagnosis on thyroid nodules through ultrasound imaging detection and fine needle aspiration cytology detection, among which about 30% cannot be determined as benign or malignant. Such patients are prone to treatment delay or over-diagnosis and treatment due to the lack of a clear treatment plan. In order to solve the problem of clinical diagnosis of thyroid cancer, in recent years, auxiliary molecular diagnostic methods represented by gene variation detection have begun to be applied in clinical practice. With the help of gene variation detection, the accuracy of clinical diagnosis has been improved, but due to the lack of clear driver gene variations in some tumors, many patients still cannot be diagnosed. Therefore, it is necessary to find a method that can further improve the accuracy of diagnosis of thyroid cancer.

[0003] More and more studies have shown that abnormal DNA methylation is an important feature of thyroid cancer. Abnormal methylation of specific genes has been found in thyroid cancer. It has been reported that hypermethylation of tumor suppressor genes (TSHR, PTEN, RASSF1A, CDKN2A, DAPK1, TIMP3, ECAD and RAP1GAP) occurs during the occurrence of thyroid cancer, and hypomethylation of MAP17, CXCL12, HORMAD2 and PFKFB2 is associated with invasive thyroid cancer. Another study found that hypermethylation of RASSF1A can be used as a biomarker for invasive tumors, and abnormal methylation of RARB2 can be used as a marker for identifying anaplastic carcinoma. These studies reveal the important role of abnormal methylation of genes in the occurrence and development of thyroid cancer and suggest that gene methylation has the potential to be used as a diagnostic marker for thyroid cancer.

[0004] The mainstream products in the North American market, Thyroseq and Afirma, are both based on genomic variation and RNA expression differences, and they perform well in terms of sensitivity and negative predictive value, but there is still room for improvement in their specificity and positive predictive value performance. A laboratory in Hope, USA, reported a set of methylation molecular markers that can be used to distinguish between benign and malignant thyroid nodules, but the prediction model constructed by it still cannot judge some samples and still has room for optimization.

[0005] Although foreign research and patents have reported that gene methylation can be used as a diagnostic marker for thyroid cancer, due to the fact that epigenetics is easily affected by the environment, there may be certain differences among different populations. So far, there are few studies on the use of methylation markers based on the Chinese population for the diagnosis of thyroid cancer, and there is no corresponding product applied in the clinic.

[0006] Therefore, the development of a methylation marker-based method for distinguishing between benign and malignant thyroid nodules in the Chinese population can help improve the accuracy of clinical diagnosis of thyroid cancer in the Chinese population, and has important application value in clinical practice. SUMMARY

[0007] The present application provides a methylation marker and use for detecting the benign and malignant thyroid nodules to solve the problems in the prior art.

[0008] One of the objectives of the present application is to provide a methylation marker for detecting the benign or malignant thyroid nodules, which comprises one or more of the following methylation regions: Chr1:12596102-12596140 region, Chr17:77830614-77830632 region, Chr1:153609274-153609317 region, Chr17:81272581-81272582 region, Chr1:1630666-1630712 region, Chr17:8151264-8151270 region, Chr1:203592359-203592399 region, Chr19:10087181-10087286 region, Chr1:21911757-21911815 region, Chr19:1253489-1253574 region, Chr1:27873182-27873192 region, Chr19:13839195-13839279 region, Chr1:3313191-3313315 region, Chr19:13846585-13846896 region, Chr1:54229814-54229861 region, Chr19:13979122-13979200 region, Chr1:58576335-58576389 region, Chr19:13986498-13986676 region, Chr1:58576421-58576603 region, Chr19:17108201-17108324 region, Chr10:110499490-110499491 region, Chr19:17292444-17292623 region, Chr10:24383461-24383566 region, Chr19:18305081-18436793 region, Chr10:75408349-75408350 region, Chr19:1851961-1852113 region, Chr10:79242311-79242381 region, Chr19:2042442-2042799 region, Chr11:10306778-10306881, Chr19:3375603-3375658 region, Chr11:110301158-110301159 region, Chr19:3434932-3434988 region, Chr11:1353646-1354505 region, Chr19:3687856-3688232 region, Chr11:19714365-19714523 region, Chr19:40806917-40807072 region,Chr 11 : 19777875-19777947, Chr 19: 46611278-46611279, Chr 11 : 46380209-46380288, Chr 19: 48480422-48480496, Chr 11 : 57426966-57427173, Chr 19: 51098902-51099067, Chr 11 : 64851310-64851314, Chr 19: 51601422-51601624, Chr 11 : 66336997-66337092, Chr 19: 52536666-52536686, Chr 11 : 70073960-70074053, Chr 19: 55087135-55087394, Chr 11 : 70742594-70742814, Chr 2: 101152526-101152688, Chr 12: 119804311-119804352, Chr 2: 237202186-237202187, Chr 12: 130980102-130980198, Chr 2: 240598627-240598628, Chr 12: 47881830-47881926, Chr 2: 25277096-25277319, Chr 12: 94644417-94644501, Chr 2: 80302861-85584386, Chr 13: 98476650-98476740, Chr 20: 22582417-22582620, Chr 14: 102928211-102928212, Chr 20: 2752143-2752688, Chr 14: 105087674-105088536, Chr 20: 57212578-57212700, Chr 15: 100115182-100115183, Chr 22: 50577975-50578304, Chr 15: 40769136-40769213, Chr 3: 65597378-65597606, Chr 16: 15143862-15144467, Chr 4: 112131117-112131180, Chr 16: 19114827-19115044, Chr 4: 153486942-153486953, Chr 16: 2124308-2176703, Chr 4: 1752763-1752832,Chr 16: 27769896-27769905, Chr 5: 139482856-139482885, Chr 16: 4943479-4943543, Chr 5: 139697245-139697591, Chr 16: 88634971-88634991, Chr 5: 141514157-141514273, Chr 15: 99123755-99123765, Chr 6: 152302126-152302178, Chr 16: 56635666-56635695, Chr 6: 165854274-165854683, Chr 16: 56648042-56648061, Chr 6: 41470838-41470918, Chr 17: 17700407-17700520, Chr 7: 1405246-1405342, Chr 17: 21464252-21464279, Chr 7: 1940287-1940414, Chr 17: 28227417-28227453, Chr 7: 256963-256964, Chr 17: 38868437-38868530, Chr 7: 74093943-74094754, Chr 17: 3963840-3963844, Chr 8: 22554216-22554275, Chr 17: 42673716-42673823, Chr 8: 27667002-27667143, Chr 17: 45143922-45144232, Chr 9: 127121856-127122337, Chr 17: 47695358-47695495, Chr 9: 131733731-131733759, Chr 17: 57875502-57875627, Chr 9: 136050771-136050820, Chr 17: 58487622-58487664, Chr 9: 136299368-136299372, Chr 17: 58487747-58488275, Chr 9: 136688112-136688211, Chr 17: 63438487-63446755, Chr 9: 137461863-137461864, Chr 17: 7174295-7174472, Chr 9: 37954187-37954325.

[0009] According to the technical scheme of the present application, the thyroid is papillary thyroid carcinoma (PTC).

[0010] According to the technical scheme of the present application, the methylation region comprises one or more of the following methylation sites: Chr1:1041251, Chr5:141514157, Chr11:1353646, Chr16:15144467, Chr19:3688090, Chr1:1630666, Chr5:141514263, Chr11:1354505, Chr16:19114827, Chr19:3688100, Chr1:1630712, Chr5:141514264, Chr11:8699310, Chr16:19115044, Chr19:3688141, Chr1:2230562, Chr5:141514273, Chr11:10306778, Chr16:25368575, Chr19:3688179, Chr1:3290597, Chr5:150297782, Chr11:10306781, Chr16:27769896, Chr19:3688232, Chr1:3313191, Chr5:150640873, Chr11:10306784, Chr16:27769904, Chr19:3695513, Chr1:3313201, Chr6:30076148, Chr11:10306824, Chr16:27769905, Chr19:4474315, Chr1:3313216, Chr6:30682948, Chr11:10306864, Chr16:28322373, Chr19:5048327, Chr1:3313227, Chr6:41470838, Chr11:10306865, Chr16:29821869, Chr19:5111340, Chr1:3313250, Chr6:41470852, Chr11:10306881, Chr16:30359428, Chr19:5325651, Chr1:3313252, Chr6:41470918, Chr11:19714365, Chr16:55331495, Chr19:5686513, Chr1:3313269, Chr6:152302126, Chr11:19714401, Chr16:56635666, Chr19:7483269, Chr1:3313271, Chr6:152302127, Chr11:19714402, Chr16:56635667, Chr19:10087181, Chr1:3313289, Chr6:152302145,Chr 11 : 19714445, Chr 16: 56635695, Chr 19: 10087286, Chr 1 : 3313292, Chr 6: 152302146, Chr 11 : 19714446, Chr 16: 56648042, Chr 19: 13839195, Chr 1 : 3313297, Chr 6: 152302157, Chr 11 : 19714514, Chr 16: 56648043, Chr 19: 13839223, Chr 1 : 3313308, Chr 6: 152302158, Chr 11 : 19714523, Chr 16: 56648061, Chr 19: 13839279, Chr 1 : 3313311, Chr 6: 152302159, Chr 11 : 19723747, Chr 16: 56662918, Chr 19: 13846585, Chr 1 : 3313315, Chr 6: 152302160, Chr 11 : 19777875, Chr 16: 66845951, Chr 19: 13846896, Chr 1 : 3505198, Chr 6: 152302167, Chr 11 : 19777883, Chr 16: 66936068, Chr 19: 13881524, Chr 1 : 3543255, Chr 6: 152302168, Chr 11 : 19777947, Chr 16: 67150229, Chr 19: 13979122, Chr 1 : 3594910, Chr 6: 152302169, Chr 11 : 46380209, Chr 16: 68371421, Chr 19: 13979200, Chr 1 : 9717812, Chr 6: 152302170, Chr 11 : 46380288, Chr 16: 72013968, Chr 19: 13986498, Chr 1 : 10635787, Chr 6: 152302171, Chr 11 : 57426966, Chr 16: 81568574, Chr 19: 13986533, Chr 1 : 12596102, Chr 6: 152302172, Chr 11 : 57427099, Chr 16: 81752390, Chr 19: 13986590, Chr 1 : 12596119, Chr 6: 152302178, Chr 11 : 57427153, Chr 16: 85436695, Chr 19: 13986591, Chr 1 : 12596120, Chr 6: 165854274, Chr 11 : 57427172, Chr 16: 88634971, Chr 19: 13986594,Chr 1 : 12596140, Chr6: 165854364, Chr11: 57427173, Chr16: 88634991, Chr19: 13986595, Chr1 : 15786745, Chr6: 165854456, Chr11: 64851310, Chr16: 88637438, Chr19: 13986598, Chr1 : 15944130, Chr6: 165854655, Chr11: 64851314, Chr17: 1937395, Chr19: 13986599, Chr1 : 16181978, Chr6: 165854683, Chr11: 64858680, Chr17: 3963840, Chr19: 13986676, Chr1 : 17540993, Chr7: 256963, Chr11: 65233327, Chr17: 3963844, Chr19: 14480374, Chr1 : 17688909, Chr7: 256964, Chr11: 66012259, Chr17: 6566846, Chr19: 17108201, Chr1 : 18927289, Chr7: 1081092, Chr11: 66336997, Chr17: 7174295, Chr19: 17108324, Chr1 : 21911757, Chr7: 1323349, Chr11: 66337010, Chr17: 7174309, Chr19: 17292444, Chr1 : 21911758, Chr7: 1405246, Chr11: 66337014, Chr17: 7174322, Chr19: 17292460, Chr1 : 21911815, Chr7: 1405342, Chr11: 66337015, Chr17: 7174352, Chr19: 17292467, Chr1 : 22318703, Chr7: 1474325, Chr11: 66337092, Chr17: 7174391, Chr19: 17292510, Chr1 : 27873182, Chr7: 1940287, Chr11: 68455664, Chr17: 7174426, Chr19: 17292547, Chr1 : 27873192, Chr7: 1940414, Chr11: 68750637, Chr17: 7174472, Chr19: 17292548, Chr1 : 41358067, Chr7: 2075109, Chr11: 68840785, Chr17: 7384019,Chr19:17292564, Chr1:54229814, Chr7:2418150, Chr11:70073960, Chr17:8151264, Chr19:17292569, Chr1:54229815, Chr7:5931371, Chr11:70074002, Chr17:8151270, Chr19:17292605, Chr1:54229860, Chr7:73904895, Chr11:70074053, Chr17:17700407, Chr19:17292623, Chr1:54229861, Chr7:74093943, Chr11:70486699, Chr17:17700408, Chr19:18305081, Chr1:58576335, Chr7:74093947, Chr11:70663065, Chr17:17700410, Chr19:18436793, Chr1:58576389, Chr7:74093981, Chr11:70742594, Chr17:17700415, Chr19:30594807, Chr1:58576421, Chr7:74094104, Chr11:70742652, Chr17:17700418, Chr19:31351829, Chr1:58576422, Chr7:74094105, Chr11:70742653, Chr17:17700425, Chr19:34577534, Chr1:58576423, Chr7:74094163, Chr11:70742654, Chr17:17700432, Chr19:35806723, Chr1:58576603, Chr7:74094245, Chr11:70742697, Chr17:17700436, Chr19:40806917, Chr1:112722970, Chr7:74094640, Chr11:70742698, Chr17:17700457, Chr19:40807049, Chr1:151838359, Chr7:74094664, Chr11:70742738, Chr17:17700468, Chr19:40807072, Chr1:151992616, Chr7:74094665, Chr11:70742739, Chr17:17700476, Chr19:46611278, Chr1:153609274, Chr7:74094687,Chr 11 : 70742755, Chr 17 : 17700516, Chr 19 : 46611279, Chr 1 : 153609317, Chr 7 : 74094688, Chr 11 : 70742756, Chr 17 : 17700520, Chr 19 : 48480422, Chr 1 : 156066660, Chr 7 : 74094725, Chr 11 : 70742776, Chr 17 : 21464252, Chr 19 : 48480496, Chr 1 : 203592359, Chr 7 : 74094726, Chr 11 : 70742786, Chr 17 : 21464269, Chr 19 : 48630748, Chr 1 : 203592399, Chr 7 : 74094732, Chr 11 : 70742814, Chr 17 : 21464279, Chr 19 : 49519137, Chr 1 : 228990544, Chr 7 : 74094733, Chr 11 : 74683980, Chr 17 : 28227417, Chr 19 : 51098902, Chr 1 : 234963449, Chr 7 : 74094754, Chr 11 : 110301158, Chr 17 : 28227423, Chr 19 : 51098906, Chr 2 : 3613345, Chr 8 : 21790103, Chr 11 : 110301159, Chr 17 : 28227453, Chr 19 : 51098914, Chr 2 : 9235477, Chr 8 : 21811216, Chr 11 : 117444044, Chr 17 : 31596788, Chr 19 : 51098924, Chr 2 : 9801337, Chr 8 : 21811216, Chr 11 : 119709773, Chr 17 : 38868437, Chr 19 : 51098928, Chr 2 : 10029534, Chr 8 : 22554216, Chr 11 : 125953540, Chr 17 : 38868530, Chr 19 : 51098930, Chr 2 : 10498150, Chr 8 : 22554275, Chr 11 : 130156804, Chr 17 : 42673716, Chr 19 : 51098945, Chr 2 : 10947098, Chr 8 : 23164199, Chr 11 : 134536759, Chr 17 : 42673728, Chr 19 : 51098973, Chr 2 : 25251595, Chr 8 : 27667002, Chr 12 : 1661591, Chr 17 : 42673757,Chr19:51098997, Chr2:25277096, Chr8:27667143, Chr12:7108181, Chr17:42673759, Chr19:51099012, Chr2:25277109, Chr8:33565861, Chr12:47881830, Chr17:42673823, Chr19:51099022, Chr2:25277110, Chr8:38651209, Chr12:47881926, Chr17:44279942, Chr19:51099032, Chr2:25277116, Chr8:38788172, Chr12:58027065, Chr17:44558988, Chr19:51099041, Chr2:25277124, Chr8:38936737, Chr12:64222777, Chr17:45143922, Chr19:51099067, Chr2:25277130, Chr8:140310313, Chr12:94150630, Chr17:45143923, Chr19:51601422, Chr2:25277131, Chr8:141308720, Chr12:94644417, Chr17:45144227, Chr19:51601423, Chr2:25277135, Chr9:35117297, Chr12:94644488, Chr17:45144232, Chr19:51601435, Chr2:25277136, Chr9:37954187, Chr12:94644501, Chr17:47695358, Chr19:51601436, Chr2:25277145, Chr9:37954267, Chr12:111443740, Chr17:47695395, Chr19:51601468, Chr2:25277319, Chr9:37954268, Chr12:117155549, Chr17:47695495, Chr19:51601469, Chr2:30230790, Chr9:37954291, Chr12:119804311, Chr17:50469031, Chr19:51601476, Chr2:33903795, Chr9:37954325, Chr12:119804352, Chr17:57868238, Chr19:51601477, Chr2:46983376, Chr9:90729572,Chr 12: 122959902, Chr 17: 57875456, Chr 19: 51601483, Chr 2: 64831509, Chr 9: 92189571, Chr 12: 124326125, Chr 17: 57875502, Chr 19: 51601484, Chr 2: 68529849, Chr 9: 101724394, Chr 12: 124928495, Chr 17: 57875544, Chr 19: 51601507, Chr 2: 74556187, Chr 9: 111851651, Chr 12: 130980102, Chr 17: 57875573, Chr 19: 51601508, Chr 2: 80302861, Chr 9: 113548848, Chr 12: 130980198, Chr 17: 57875612, Chr 19: 51601509, Chr 2: 85584290, Chr 9: 114108496, Chr 13: 34750527, Chr 17: 57875627, Chr 19: 51601510, Chr 2: 85584348, Chr 9: 121282620, Chr 13: 80490728, Chr 17: 58487622, Chr 19: 51601580, Chr 2: 85584364, Chr 9: 124811105, Chr 13: 98476650, Chr 17: 58487643, Chr 19: 51601581, Chr 2: 85584373, Chr 9: 125577538, Chr 13: 98476685, Chr 17: 58487664, Chr 19: 51601597, Chr 2: 85584379, Chr 9: 127121856, Chr 13: 98476703, Chr 17: 58487747, Chr 19: 51601603, Chr 2: 85584386, Chr 9: 127121880, Chr 13: 98476740, Chr 17: 58487814, Chr 19: 51601624, Chr 2: 101152526, Chr 9: 127121963, Chr 14: 23115298, Chr 17: 58488080, Chr 19: 52536666, Chr 2: 101152638, Chr 9: 127121964, Chr 14: 38256108, Chr 17: 58488116, Chr 19: 52536686, Chr 2: 101152643, Chr 9: 127121994, Chr 14: 64743251, Chr 17: 58488275,Chr19:55087135, Chr2:101152644, Chr9:127122016, Chr14:68695352, Chr17:59246476, Chr19:55087177, Chr2:101152688, Chr9:127122021, Chr14:73718571, Chr17:63438487, Chr19:55087394, Chr2:113251274, Chr9:127122092, Chr14:91375612, Chr17:63446755, Chr20:2752143, Chr2:127064916, Chr9:127122093, Chr14:94291257, Chr17:74852370, Chr20:2752154, Chr2:219252952, Chr9:127122165, Chr14:101621513, Chr17:76140257, Chr20:2752233, Chr2:236643030, Chr9:127122179, Chr14:102928211, Chr17:77409611, Chr20:2752646, Chr2:237202186, Chr9:127122257, Chr14:102928212, Chr17:77830614, Chr20:2752655, Chr2:237202187, Chr9:127122258, Chr14:103135447, Chr17:77830632, Chr20:2752673, Chr2:240598627, Chr9:127122266, Chr14:103932682, Chr17:78130182, Chr20:2752688, Chr2:240598628, Chr9:127122267, Chr14:104790584, Chr17:78884051, Chr20:20887989, Chr3:19234357, Chr9:127122297, Chr14:104978811, Chr17:80892292, Chr20:22582417, Chr3:43994917, Chr9:127122337, Chr14:105087674, Chr17:81272581, Chr20:22582495, Chr3:47542858, Chr9:131380762, Chr14:105087732, Chr17:81272582, Chr20:22582508,Chr3:49685862, Chr9:131733731, Chr14:105087733, Chr17:81401923, Chr20:22582551, Chr3:50322471, Chr9:131733759, Chr14:105087752, Chr17:82807710, Chr20:22582552, Chr3:65597378, Chr9:133772391, Chr14:105087766, Chr18:2898307, Chr20:22582556, Chr3:65597401, Chr9:136050771, Chr14:105087773, Chr18:10454023, Chr20:22582560, Chr3:65597443, Chr9:136050820, Chr14:105087774, Chr18:47531839, Chr20:22582620, Chr3:65597606, Chr9:136299368, Chr14:105087820, Chr18:75089977, Chr20:23656139, Chr3:94044263, Chr9:136299372, Chr14:105087821, Chr19:1016182, Chr20:33644652, Chr3:100399111, Chr9:136588605, Chr14:105087825, Chr19:1046455, Chr20:36808863, Chr3:107430527, Chr9:136688112, Chr14:105087826, Chr19:1136348, Chr20:40690738, Chr3:138071317, Chr9:136688211, Chr14:105087827, Chr19:1253489, Chr20:45373105, Chr3:146497401, Chr9:136997998, Chr14:105087828, Chr19:1253490, Chr20:50865778, Chr3:180829739, Chr9:137461863, Chr14:105087829, Chr19:1253574, Chr20:51487679, Chr3:194277184, Chr9:137461864, Chr14:105087841, Chr19:1257191, Chr20:57212578, Chr3:194397453,Chr 10: 1516171, Chr 14: 105087842, Chr 19: 1387756, Chr 20: 57212681, Chr 3: 194487586, Chr 10: 3780554, Chr 14: 105087845, Chr 19: 1851961, Chr 20: 57212686, Chr 4: 1023408, Chr 10: 17287619, Chr 14: 105088536, Chr 19: 1852113, Chr 20: 57212687, Chr 4: 1752763, Chr 10: 24383461, Chr 14: 105489404, Chr 19: 1860626, Chr 20: 57212700, Chr 4: 1752824, Chr 10: 24383504, Chr 15: 31393344, Chr 19: 2042442, Chr 20: 58613587, Chr 4: 1752832, Chr 10: 24383566, Chr 15: 40769136, Chr 19: 2042495, Chr 20: 61698174, Chr 4: 2270694, Chr 10: 44293658, Chr 15: 40769163, Chr 19: 2042544, Chr 20: 62259203, Chr 4: 2801347, Chr 10: 71260292, Chr 15: 40769213, Chr 19: 2042594, Chr 20: 62416735, Chr 4: 7740007, Chr 10: 71715990, Chr 15: 43799969, Chr 19: 2042637, Chr 20: 63562476, Chr 4: 8156798, Chr 10: 75399202, Chr 15: 55492672, Chr 19: 2042799, Chr 20: 63978593, Chr 4: 8229022, Chr 10: 75408349, Chr 15: 70349077, Chr 19: 2207249, Chr 21: 31539560, Chr 4: 112131117, Chr 10: 75408350, Chr 15: 72227398, Chr 19: 3208942, Chr 21: 43194186, Chr 4: 112131180, Chr 10: 77636464, Chr 15: 77887216, Chr 19: 3375603, Chr 21: 44653329, Chr 4: 121711736, Chr 10: 79080072, Chr 15: 88639608, Chr 19: 3375604,Chr21 :45430879, Chr4:153486942, Chr10:79113409, Chr15:99123755, Chr19:3375607, Chr22:23457136, Chr4:153486953, Chr10:79242311, Chr15:99123765, Chr19:3375617, Chr22:30224646, Chr4:185741527, Chr10:79242326, Chr15:100115182, Chr19:3375627, Chr22:36577478, Chr5:5495050, Chr10:79242354, Chr15:100115183, Chr19:3375658, Chr22:37676630, Chr5:14577348, Chr10:79242360, Chr16:628678, Chr19:3434932, Chr22:38294213, Chr5:55221002, Chr10:79242372, Chr16:785820, Chr19:3434965, Chr22:39292621, Chr5:139482856, Chr10:79242373, Chr16:1303808, Chr19:3434987, Chr22:43244984, Chr5:139482885, Chr10:79242377, Chr16:1379904, Chr19:3434988, Chr22:44625144, Chr5:139697245, Chr10:79242380, Chr16:1964269, Chr19:3458219, Chr22:44841705, Chr5:139697251, Chr10:79242381, Chr16:2124308, Chr19:3687856, Chr22:45537390, Chr5:139697256, Chr10:86224885, Chr16:2124309, Chr19:3687881, Chr22:45922257, Chr5:139697265, Chr10:97922461, Chr16:2124359, Chr19:3687882, Chr22:45996819, Chr5:139697304, Chr10:100371395, Chr16:2176703, Chr19:3687924, Chr22:46006251, Chr5:139697306, Chr10:110499490,Chr 16: 2929683, Chr 19: 3687984, Chr 22: 46535376, Chr 5: 139697317, Chr 10: 110499491, Chr 16: 4943479, Chr 19: 3688013, Chr 22: 49600835, Chr 5: 139697446, Chr 10: 114631841, Chr 16: 4943481, Chr 19: 3688023, Chr 22: 49959739, Chr 5: 139697484, Chr 10: 120456377, Chr 16: 4943543, Chr 19: 3688033, Chr 22: 50197602, Chr 5: 139697498, Chr 10: 124094248, Chr 16: 10877393, Chr 19: 3688041, Chr 22: 50577975, Chr 5: 139697550, Chr 10: 127907972 Chr 16: 11182276, Chr 19: 3688042, Chr 22: 50578304, Chr 5: 139697582, Chr 10: 132699462, Chr 16: 11669773, Chr 19: 3688069, Chr 22: 50704114, Chr 5: 139697591, Chr 11: 627379, Chr 16: 15143862, Chr 19: 3688070, Chr X: 132420212.

[0011] According to the technical scheme of the present application, the methylation markers include one or more of the following methylation sites: Chr20:22582560, Chr14:64743251, Chr9:133772391, Chr22:50704114, Chr7:2418150, Chr10:79080072, Chr2:85584290, Chr20:62259203, Chr11:70486699, Chr22:36577478, Chr16:1964269, Chr22:49600835, Chr17:76140257, Chr1:1041251, Chr20:33644652, Chr9:127122179, Chr16:66845951, Chr7:256964, Chr16:66936068, Chr20:61698174, Chr14:103932682, Chr19:3375607, Chr19:10087286, Chr20:50865778, Chr20:22582508, Chr3:146497401, Chr5:139697251, Chr20:62416735, Chr11:70742739, Chr4:8156798.

[0012] According to the technical scheme of the present application, the methylation markers include one or more of the following methylation sites: Chr20:22582560, Chr17:76140257, Chr14:64743251, Chr1:1041251, Chr7:2418150, Chr16:66845951, Chr10:79080072, Chr7:256964, Chr11:70486699, Chr14:103932682, Chr22:36577478, Chr19:3375607.

[0013] The second object of the present application is to provide use of a substance for detecting the methylation markers as described above in the preparation of a product for detecting the benign or malignant thyroid nodules.

[0014] According to the technical scheme of the present application, the product includes a substance for detecting the methylation level of the methylation markers as described above.

[0015] Preferably, the substance for detecting the methylation marker is selected from one or more of primers, probes, PCR buffer, polymerase, dNTPs, bisulfite and its derivatives, methylation-sensitive or -insensitive restriction enzymes, restriction enzyme buffer, fluorescent dyes, fluorescence quenchers, fluorescence reporters, exonucleases, alkaline phosphatase, internal standards and controls. Among them, the primers and probes are used to detect the methylation level of the methylation marker.

[0016] Preferably, the methylation level is obtained by one or more of reduced representation bisulfite sequencing (RRBS), methylation-specific PCR (MSP), methylation-specific real-time PCR (MSRE-qPCR), whole genome bisulfite sequencing (WGBS) and methylation chip. In a certain preferred embodiment of the present application, it is obtained by reduced representation bisulfite sequencing (RRBS).

[0017] The third object of the present application is to provide a product for detecting the benign or malignant thyroid nodule, which comprises a substance for detecting the methylation marker as described above.

[0018] According to the technical solution of the present application, the product comprises at least one of a kit, a chip, a probe or a membrane strip.

[0019] The fourth object of the present application is to provide a method for constructing a thyroid nodule benign or malignant evaluation model, which comprises the following steps:

[0020] 1) Constructing a sample data set based on the methylation level of the methylation marker as described above, sample information and benign or malignant information in the biological sample of a subject, wherein the subject comprises a patient with a diagnosed benign thyroid nodule and a patient with a thyroid malignant nodule, different classification codes are assigned to the malignant sample and the benign sample, and a classification threshold is set based on the classification codes;

[0021] 2) Randomly dividing the sample data set into a test set and a training set, learning the data of the training set using a machine learning method, and obtaining the thyroid nodule benign or malignant evaluation model.

[0022] According to the technical solution of the present application, the biological sample is selected from at least one of plasma, serum, whole blood, urine, amniotic fluid and tissue.

[0023] According to the technical solution of the present application, the machine learning method is selected from at least one of support vector machine, random forest algorithm and neural network algorithm.

[0024] According to the technical solution of the present application, the division ratio of the test set and the training set is (5-7) (3-5).

[0025] According to the technical scheme of the present application, the thyroid nodule benign and malignant evaluation model is used for outputting a prediction value, when the prediction value is higher than the classification threshold value, then it is evaluated as a thyroid malignant nodule; when the prediction value is lower than the classification threshold value, then it is evaluated as a thyroid benign nodule.

[0026] According to the technical scheme of the present application, it further comprises a step 3) of verifying the thyroid nodule benign and malignant evaluation model by using a test set.

[0027] The fifth object of the present application is to provide a device for constructing a thyroid nodule benign and malignant evaluation model, which comprises:

[0028] A data collection module is used for constructing a sample data set based on the methylation level of the methylation marker in the biological sample of the subject as described above, wherein the subject comprises a patient with a diagnosed thyroid benign nodule and a patient with a thyroid malignant nodule, different classification codes are given to the malignant sample and the benign sample, and a classification threshold value is set based on the classification code;

[0029] A model construction module is used for randomly dividing the sample data set into a training set and a verification set, and learning the training set by using a machine learning method to obtain the thyroid nodule benign and malignant evaluation model.

[0030] According to the technical scheme of the present application, the biological sample is selected from at least one of plasma, serum, whole blood, urine, amniotic fluid and tissue.

[0031] According to the technical scheme of the present application, the machine learning method is selected from at least one of a support vector machine, a random forest algorithm and a neural network algorithm. In some preferred embodiments of the present application, it is a random forest model.

[0032] According to the technical scheme of the present application, the division ratio of the test set and the training set is (5-7) (3-5).

[0033] According to the technical scheme of the present application, the thyroid nodule benign and malignant evaluation model is used for outputting a prediction value, when the prediction value is higher than the classification threshold value, then it is evaluated as a thyroid malignant nodule; when the prediction value is lower than the classification threshold value, then it is evaluated as a thyroid benign nodule.

[0034] Preferably, the classification threshold value is 0.5.

[0035] Preferably, the code of the malignant sample is 1, and the code of the benign sample is 0.

[0036] The sixth object of the present application is to provide a device for evaluating the benign and malignant of thyroid nodules, which comprises:

[0037] an input module configured to obtain data of the detection amount of the methylation marker in the biological sample of the subject;

[0038] a benign-malignant evaluation module configured to calculate the data of the detection amount obtained by the input module using the benign-malignant evaluation model of the thyroid nodule obtained by the construction method, to obtain a prediction value, and to compare the prediction value with a classification threshold to obtain a benign-malignant result of the thyroid nodule;

[0039] an output module configured to output the benign-malignant result of the thyroid nodule.

[0040] According to the technical solution of the present application, the benign-malignant evaluation model of the thyroid nodule is used to output a prediction value, and when the prediction value is higher than the classification threshold, the thyroid nodule is evaluated as malignant; and when the prediction value is lower than the classification threshold, the thyroid nodule is evaluated as benign.

[0041] Preferably, the classification threshold is 0.5.

[0042] The seventh object of the present application is to provide a method for detecting the benign-malignant risk of a thyroid nodule, comprising the following steps:

[0043] S1, obtaining data of the detection amount of the methylation marker in the biological sample of the subject;

[0044] S2, calculating the data of the detection amount using the benign-malignant evaluation model of the thyroid nodule obtained by the construction method, to obtain a prediction value, and comparing the prediction value with a classification threshold to obtain a benign-malignant result of the thyroid nodule;

[0045] S3, according to the calculation and comparison result, outputting the benign-malignant result of the thyroid nodule.

[0046] According to the technical solution of the present application, in step S1, the detection amount of the methylation marker is the methylation level of the marker, which is determined by quantitative analysis of the genomic DNA in the biological sample.

[0047] Preferably, the methylation level is obtained by one or more of reduced representation bisulfite sequencing (RRBS), methylation-specific PCR (MSP), methylation-specific real-time PCR (MSRE-qPCR), whole-genome bisulfite sequencing (WGBS), and methylation chip. In a preferred embodiment of the present application, it is obtained by reduced representation bisulfite sequencing (RRBS).

[0048] According to the technical scheme of the present application, in step S1, the biological sample is selected from at least one of plasma, serum, whole blood, urine, amniotic fluid and tissue. As a preferred embodiment, the tissue is selected from a fresh tissue sample, a fine needle biopsy sample or a formalin-fixed paraffin-embedded (FFPE) sample of a thyroid nodule of the subject. In a preferred embodiment of the present application, the fresh tissue sample is of a thyroid nodule.

[0049] An eighth object of the present application is to provide a device comprising a processor and a memory for storing a computer program, the processor being configured to execute the computer program stored in the memory to cause the device to perform the construction method or the detection method as described above.

[0050] A ninth object of the present application is to provide a computer-readable storage medium having stored thereon a computer program, the program being configured to, when executed by a processor, implement the construction method or the detection method as described above.

[0051] Advantages of the present application:

[0052] 1) The methylation markers for detecting the benign and malignant thyroid nodules of the present application are screened based on the data of thyroid malignant nodules and benign nodules, can effectively reflect the differences in gene methylation levels between benign and malignant nodules, have good distinguishing effect on benign and malignant nodules, and are more in line with the actual clinical diagnosis compared with the thyroid malignant nodules and the adjacent tissues commonly used in other studies, and have higher clinical practical value.

[0053] 2) The present application performs high-throughput screening and identifies 780 new methylation sites, associates 273 tumor occurrence and development related genes, involves Fc gamma R-mediated phagocytosis signal pathway, and provides methylation markers for detecting the benign and malignant thyroid nodules, which contain multiple genes and multiple methylation sites. Through the detection of the methylation levels of multiple genes and multiple sites, the problems of low and unstable methylation signals of single gene or single site are overcome, thereby improving the sensitivity and specificity of the detection.

[0054] 3) The methylation markers and the thyroid nodule benign and malignant evaluation model provided by the present application can be used for the auxiliary diagnosis of the benign and malignant thyroid nodules. The accuracy of the model is 95%, the AUC is 100%, the sensitivity is 100%, and the specificity is 91%. The model has good prediction effect on the benign and malignant thyroid nodules, and can improve the accuracy of the clinical diagnosis of the benign and malignant thyroid nodules.

[0055] 4) The thyroid benign and malignant evaluation model of the application not only has extremely high sensitivity, helps to reduce missed diagnosis, is conducive to early treatment, and improves patient survival rate; and it has a relatively high negative predictive value, can avoid unnecessary surgery for patients, reduce social medical and economic burden, and improve the overall survival quality of thyroid nodule patients.

[0056] 5) The thyroid benign and malignant detection model provided by the application can directly give a benign and malignant prediction result by analyzing the methylation level of the methylation marker in the biological sample to be detected, avoiding the subjectivity of result interpretation in ultrasonic imaging detection and cytopathology detection, improving the consistency and accuracy of result interpretation, and facilitating clinical popularization and application.

[0057] 6) In clinical practice, the method of the application can be used as an auxiliary traditional diagnostic means in combination with fine needle aspiration cytology detection, without the need for additional sampling, simplifying the diagnosis process, improving detection efficiency, and reducing the pain of patients.

[0058] 7) The application is developed based on clinical data of Chinese population, and the identified methylation markers and the constructed thyroid benign and malignant evaluation model are more suitable for the benign and malignant evaluation of thyroid nodules of Chinese population. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 Important features of the thyroid benign and malignant evaluation model in Example 1 of the application are shown.

[0060] Figure 2 The ROC curve of the thyroid benign and malignant evaluation model in Example 1 of the application is shown. DETAILED DESCRIPTION

[0061] The embodiments of the application will be described in detail with specific reference being made to certain particular embodiments of the application. The advantages of the application will be apparent from the following description of the embodiments of the application. The following description of the embodiments of the application is made with reference to the accompanying drawings. The application can be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art. In the drawings, like reference numerals refer to like elements throughout the several views.

[0062] Before further description of the specific embodiments of the application, it is to be understood that the application is not limited to the particular specific embodiments described below; it is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; in this specification and in the claims, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise.

[0063] When the embodiments give numerical ranges, it is understood that, unless the present invention indicates otherwise, each numerical range's two endpoints, and any number in between the two endpoints, can be selected. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by one of ordinary skill in the art. Except in the Examples, or where otherwise explicitly indicated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Unless otherwise specified, any method, device, material or ingredient similar or equivalent to those described in the present invention can be used in the practice of the present invention.

[0064] The first aspect of the present invention protects a methylation marker for detecting the benign or malignant of thyroid nodule, comprising one or more of the following methylation regions, based on the sequence of human reference genome Hg38. The methylation regions are shown in Table 1.

[0065] In the present invention, the thyroid is preferably papillary thyroid carcinoma (PTC).

[0066] Table 1

[0067]

[0068]

[0069] In the present invention, the methylation region comprises one or more of the following methylation sites, the methylation sites are shown in Table 2.

[0070] Table 2

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] In the present invention, as a preferred embodiment, the methylation marker comprises one or more of the following methylation sites. The methylation sites are shown in Table 3.

[0077] Table 3

[0078] Chr20:22582560 Chr14:64743251 Chr9:133772391 Chr22:50704114 Chr7:2418150 Chr10:79080072 Chr2:85584290 Chr20:62259203 Chr11:70486699 Chr22:36577478 Chr16:1964269 Chr22:49600835 Chr17:76140257 Chr1:1041251 Chr20:33644652 Chr9:127122179 Chr16:66845951 Chr7:256964 Chr16:66936068 Chr20:61698174 Chr14:103932682 Chr19:3375607 Chr19:10087286 Chr20:50865778 Chr20:22582508 Chr3:146497401 Chr5:139697251 Chr20:62416735 Chr11:70742739 Chr4:8156798

[0079] In the present invention, as a preferred embodiment, the methylation marker comprises one or more of the following methylation sites. The methylation sites are shown in Table 4.

[0080] Table 4

[0081] Chr20:22582560 Chr17:76140257 Chr14:64743251 Chr1:1041251 Chr7:2418150 Chr16:66845951 Chr10:79080072 Chr7:256964 Chr11:70486699 Chr14:103932682 Chr22:36577478 Chr19:3375607

[0082] Another aspect of the present application protects a gene associated with the methylation markers described above, the gene being selected from at least one of the following, the gene being seen in Table 5.

[0083] Table 5

[0084]

[0085]

[0086] Another aspect of the present application protects the use of a substance for detecting the methylation markers described above in the preparation of a product for detecting the benign or malignant nature of thyroid nodules.

[0087] In the present application, the substance for detecting the methylation markers is selected from one or more of primers, probes, PCR buffers, polymerases, dNTPs, bisulfite and its derivatives, methylation-sensitive or -insensitive restriction enzymes, restriction enzyme buffers, fluorescent dyes, fluorescence quenchers, fluorescence reporters, exonucleases, alkaline phosphatase, internal standards and controls. Among them, the primers and probes are used to detect the methylation level of the methylation markers.

[0088] In the present application, the product includes a substance for detecting the methylation level of the methylation markers described above.

[0089] In the present application, the method for detecting the methylation level of the methylation markers is selected from one or more of reduced representation bisulfite sequencing (RRBS), methylation-specific PCR (MSP), quantitative real-time PCR (MSRE-qPCR), whole-genome bisulfite sequencing (WGBS) and methylation chip.

[0090] Another aspect of the present application protects a product for detecting the benign or malignant nature of thyroid nodules, the product comprising a substance for detecting the methylation markers described above.

[0091] In the present application, the product includes at least one of a kit, a chip, a probe or a membrane strip.

[0092] Another aspect of the present application protects a method for constructing a model for evaluating the benign or malignant nature of thyroid nodules, the method comprising the following:

[0093] 1) based on the methylation level of the methylation marker in the biological sample of the subject as described above, sample information and benign and malignant information, a sample data set is constructed, wherein the subjects include patients with diagnosed benign thyroid nodules and patients with malignant thyroid nodules, different classification codes are assigned to malignant samples and benign samples, and a classification threshold is set based on the classification codes;

[0094] 2) the sample data set is randomly divided into a test set and a training set, and a machine learning method is used to learn the data of the training set to obtain the thyroid nodule benign and malignant evaluation model.

[0095] In the present application, the machine learning method is selected from at least one of support vector machine, random forest algorithm and neural network algorithm. In some preferred embodiments of the present application, it is a random forest model.

[0096] In the present application, the division ratio of the test set and the training set is (5-7) (3-5). In some preferred embodiments of the present application, the division ratio of the test set and the training set is 6:4.

[0097] In the present application, the thyroid nodule benign and malignant evaluation model is used to output a prediction value, and when the prediction value is higher than the classification threshold, it is evaluated as a malignant thyroid nodule; when the prediction value is lower than the classification threshold, it is evaluated as a benign thyroid nodule.

[0098] In the present application, the classification threshold is 0.5.

[0099] In the present application, the classification code of the malignant sample is 1, and the classification code of the benign sample is 0.

[0100] In the present application, the classification threshold is set according to the diagnostic data, for example, when modeling, the classification code of the malignant sample is 1, and the classification code of the benign sample is 0. During the process of training the model by R software and random forest, the classification threshold is set to 0.5 by default. The model constructed finally also takes 0.5 as the classification threshold to distinguish between benign and malignant for the subjects.

[0101] In the present application, it also includes step 3) verifying the thyroid nodule benign and malignant evaluation model by using the test set.

[0102] Another aspect of the present application protects a device for constructing a thyroid nodule benign and malignant evaluation model, which comprises:

[0103] A data collection module is used to construct a sample data set based on the methylation level of the methylation marker in the biological sample of the subject as described above, wherein the subjects include patients with diagnosed benign thyroid nodules and patients with malignant thyroid nodules, different classification codes are assigned to malignant samples and benign samples, and a classification threshold is set based on the classification codes;

[0104] a model construction module, configured to randomly divide the sample dataset into a training set and a test set, and learn the training set using a machine learning method to obtain the thyroid nodule benignity evaluation model.

[0105] In the present application, the machine learning method is selected from at least one of a support vector machine, a random forest algorithm and a neural network algorithm. In some preferred embodiments of the present application, the random forest model is used.

[0106] In the present application, the division ratio of the test set and the training set is (5-7):(3-5). In some preferred embodiments of the present application, the division ratio of the test set and the training set is 6:4.

[0107] In the present application, the thyroid nodule benignity evaluation model is used to output a prediction value, and when the prediction value is higher than the classification threshold, it is evaluated as a thyroid malignant nodule; when the prediction value is lower than the classification threshold, it is evaluated as a thyroid benign nodule.

[0108] In the present application, the classification threshold is 0.5.

[0109] In the present application, the classification code of the malignant sample is 1, and the classification code of the benign sample is 0.

[0110] In the present application, the classification threshold is set according to the diagnostic data, for example, when modeling, the classification code of the malignant sample is 1, and the classification code of the benign sample is 0. During the process of training the model by R software and random forest, the classification threshold is set to 0.5 by default. The finally constructed model also takes 0.5 as the classification threshold to distinguish between benign and malignant for the subjects.

[0111] Another aspect of the present application protects a device for evaluating the benignity and malignancy of thyroid nodules, which comprises:

[0112] an input module for obtaining the detection amount data of the methylation markers in the biological sample of the subject as described above;

[0113] a benignity and malignancy evaluation module for calculating the detection amount data of the input module using the thyroid nodule benignity evaluation model obtained by the construction method described above to obtain a prediction value, and comparing the prediction value with a classification threshold to obtain the benignity and malignancy result of the thyroid nodule;

[0114] an output module for outputting the benignity and malignancy result of the thyroid nodule.

[0115] In the present application, when the prediction value is higher than the classification threshold, it is evaluated as a thyroid malignant nodule; when the prediction value is lower than the classification threshold, it is evaluated as a thyroid benign nodule.

[0116] In the present application, the classification threshold is 0.5.

[0117] Another aspect of the present application protects a method for detecting the risk of benign or malignant thyroid nodules, comprising the following steps:

[0118] S1, obtaining the detection amount data of the methylation markers in the biological sample of the subject to be tested as described above;

[0119] S2, applying the thyroid nodule benign or malignant evaluation model obtained by the construction method as described above to calculate the detection amount data, obtaining a predicted value, and comparing it with the classification threshold to obtain the benign or malignant result of the thyroid nodule;

[0120] S3, according to the calculation and comparison result, identifying and outputting the benign or malignant result of the thyroid nodule.

[0121] In the present application, in step S1, the detection amount is the methylation level of the methylation marker, which is determined by quantitative analysis of genomic DNA in the biological sample.

[0122] In the present application, in step S2, when the predicted value is higher than the classification threshold, it is evaluated as a malignant thyroid nodule; when the predicted value is lower than the classification threshold, it is evaluated as a benign thyroid nodule.

[0123] Another aspect of the present application protects a device comprising a processor and a memory, the memory being used to store a computer program, and the processor being used to execute the computer program stored in the memory to make the device execute the construction method as described above or the detection method as described above.

[0124] Another aspect of the present application protects a computer readable storage medium having a computer program stored thereon, which program, when executed by a processor, implements the construction method as described above or the detection method as described above.

[0125] In the present application, the biological sample is selected from at least one of plasma, serum, whole blood, urine, amniotic fluid and tissue. As a preferred embodiment, the tissue is selected from a fresh tissue sample, a fine needle biopsy sample or a formalin-fixed paraffin-embedded (FFPE) sample of a thyroid nodule of a subject. In a preferred embodiment of the present application, it is a fresh tissue sample of a thyroid nodule.

[0126] In the present application, the detection method of the methylation level of the methylation marker is selected from one or more of reduced representation bisulfite sequencing (RRBS), methylation-specific PCR (MSP), fluorescent quantitative PCR (MSRE-qPCR), whole genome bisulfite sequencing (WGBS) and methylation chip.

[0127] Preferably, reduced representation bisulfite sequencing (RRBS) is to use methylation-insensitive restriction enzyme Mspl to treat DNA; to perform end repair, A-tailing and ligation of sequencing adapters on all cytosine modified by methylation on the DNA fragments after enzyme treatment; to recover DNA fragments with a length of 40-220 bp by gel electrophoresis; then to perform bisulfite treatment (using EZ DNA Methylation Gold Kit, Zymo Research), after treatment, unmethylated C becomes U (T after PCR amplification), and methylated C remains unchanged, and finally to perform PCR amplification to obtain the final DNA library and sequencing, and then to statistically analyze the data to obtain the methylation level of genomic DNA.

[0128] Preferably, methylation-specific PCR (MSP) is based on sodium bisulfite treatment of genomic DNA, unmethylated cytosine becomes uracil, and methylated cytosine remains unchanged, different specific primers are designed to amplify the DNA methylation region of interest by PCR, and the product is sequenced by Sanger or agarose gel electrophoresis to determine the methylation state of the DNA sequence complementary to the primer. Preferably, methylation-sensitive restriction enzyme-dependent quantitative PCR (MSRE-qPCR) is based on the use of methylation-sensitive restriction enzymes to treat genomic DNA, unmethylated DNA sequences are cut, and methylated DNA sequences remain intact, specific primers and probes are used to amplify the target region of interest, and methylated intact DNA fragments are successfully amplified. The region is not methylated and cannot be amplified due to being cut into small fragments by the restriction enzyme, and finally the methylation state of the DNA sequence is determined by software analysis according to the fluorescence signal intensity captured by the instrument.

[0129] Preferably, whole genome bisulfite sequencing (WGBS) first fragments the genome, then uses sodium bisulfite to treat DNA to cause unmethylated C to become U and become T in subsequent PCR and sequencing, while methylated C is not affected, then performs DNA repair, adds adapters and performs PCR amplification to obtain the final DNA library and sequencing, then statistically analyzes the data to obtain the methylation level of genomic DNA.

[0130] Preferably, methylation chip technology is based on the design of probes for the target DNA region, and the proportion of methylated CpG sites and non-methylated CpG sites is distinguished by the signals of two different probes, and the methylation proportion of the target site is calculated according to the values of the two fluorescence signals. The one currently used on the market is the 850K chip.

[0131] Analysis of DNA methylation level of thyroid nodule in Example 1

[0132] In this example, patients with malignant thyroid nodules and patients with benign thyroid nodules were taken as the research objects, genomic DNA was extracted, and RRBS was used for sequencing of the extracted genomic DNA to obtain the methylation level of each C site of the genomic DNA of patients with malignant thyroid nodules and patients with benign thyroid nodules. It includes the following steps:

[0133] 1.1 Research object

[0134] 30 patients with malignant thyroid nodules and 19 patients with benign thyroid nodules, patients with malignant thyroid nodules have papillary carcinoma (PTC).

[0135] 2.2 DNA extraction

[0136] Obtain the thyroid nodule tissue samples of 49 patients in step 1.1, and extract genomic DNA using QIAamp DNA Mini Kit. Including the following:

[0137] (1) Cut a small piece of tissue into a 1.5 mL centrifuge tube, add lysis buffer and proteinase K, mix well, and incubate at 56°C until the tissue is completely lysed.

[0138] (2) Add RNAse A, mix well, and incubate at room temperature for 2 minutes.

[0139] (3) Add stop buffer, mix well, and incubate at 70°C for 10 minutes.

[0140] (4) Add anhydrous ethanol, mix well, centrifuge briefly, transfer to the collection column, centrifuge for 1 minute, and discard the waste liquid.

[0141] (5) Add wash buffer, centrifuge for 1 minute, and discard the waste liquid.

[0142] (6) Add wash buffer again, centrifuge for 3 minutes, and discard the waste liquid.

[0143] (7) Place the collection column at the highest speed for 1 minute, discard the remaining waste liquid.

[0144] (8) Add elution buffer to the center of the membrane of the collection column, incubate at room temperature for 1 minute, centrifuge for 1 minute, and elute the DNA.

[0145] (9) Use agarose gel electrophoresis to analyze the degree of DNA degradation and whether there is RNA contamination. The DNA should not be significantly degraded and contaminated; use Qubit fluorescence quantifier to determine the DNA concentration. The total amount of DNA should not be less than 1.5 μg, which is qualified DNA.

[0146] 1.3 Screening of methylation markers

[0147] 1) Simplify the construction of the library of methylation sequencing (RRBS)

[0148] (1) Add 1% negative control (lambda DNA) to the qualified DNA obtained in step 1.2.

[0149] (2) Use methylation-insensitive restriction endonuclease MspI to perform enzyme treatment on the DNA.

[0150] (3) Perform end repair and A-tailing on the enzyme-treated DNA fragments to obtain A-tailed DNA fragments.

[0151] (4) Connect sequencing adapters with methylated cytosine to both ends of the A-tailed DNA fragments in step (3).

[0152] (5) Perform agarose gel electrophoresis on the DNA fragments with sequencing adapters in step (4), and recover the DNA fragments with a length of 40-220 bp by cutting the gel.

[0153] (6) Use EZ DNA Methylation Gold Kit (Zymo Research) to perform bisulfite treatment on the DNA fragments in step (5). After treatment, unmethylated cytosine (C) is converted to uracil (U) (which is then changed to thymine (T) after PCR amplification), while methylated cytosine (C) remains unchanged.

[0154] (7) Perform PCR amplification on the treated DNA in step (6) to obtain the final DNA library.

[0155] 2) Library quality control

[0156] (1) Use Qubit fluorometer to preliminarily quantify the DNA library in step 1), and dilute the library to 1 ng / μL.

[0157] (2) Use Agilent 2100 Bioanalyzer to detect the insert size of the library, with the main peak at 250-350 bp.

[0158] (3) Use qPCR to accurately quantify the effective concentration of the library, and the effective concentration of the qualified library should be greater than 2 nM.

[0159] 3) Library sequencing

[0160] The qualified library is mixed according to the effective concentration and the target data volume, and then sequenced on the Illumina sequencing platform to obtain the sequencing results.

[0161] 4) Methylation site analysis

[0162] (1) The sequencing results obtained in step 3) were subjected to quality assessment using FastQC, and then subjected to quality control filtration using fastp.

[0163] (2) Reference genome alignment analysis was performed using Bismark: first, Bismark converted both the quality-filtered sequencing reads and the reference genome sequence from C to T and G to A (reverse complement sequence), then the converted sequencing reads and the converted genome sequence were pairwise aligned, and the best alignment result among the four parallel alignment results was taken as the final alignment result. The methylation status of all C sites was counted.

[0164] In addition, the percentage of C was obtained by sequencing the C sites in the lambda DNA reference sequence, and the bisulfite conversion rate (r) was calculated.

[0165] (3) Calculate the methylation level: for each C site, count the number of reads supporting methylation of the site (mC) and the number of reads supporting non-methylation of the site (umC).

[0166] Based on the number of methylation reads (mC), the total number of reads (mC+umC), and the conversion rate (r) of C sites, a binomial distribution test was performed, and sites with FDR-adjusted p-value <0.05 were defined as methylation sites.

[0167] The methylation level of each site was calculated by counting the number of methylation and non-methylation reads covering each C site, and calculating the methylation level (ML) of each C site, the calculation formula is as follows:

[0168] ML = mC / (mC+umC).

[0169] (4) Obtain the methylation level of each C site of the genomic DNA of malignant and benign thyroid nodules.

[0170] Example 2 Screening of methylation markers and construction of thyroid benign and malignant evaluation model

[0171] In this embodiment, the methylation level of each C site of the genomic DNA of the patients with malignant thyroid nodules and the patients with benign thyroid nodules obtained in Example 1 was screened to obtain methylation markers for detecting the benign and malignant thyroid nodules, and a thyroid benign and malignant evaluation model was constructed according to the screened methylation markers. It includes the following:

[0172] 2.1 Analysis of differential methylation sites to obtain methylation markers for detecting benign and malignant thyroid nodules

[0173] Differential methylation analysis was performed using the methylKit (v1.12.0) from the R (V3.6.2) package, including the following steps:

[0174] 1) First, filter out sites with a coverage of less than 10 and sites with a coverage of more than 99.9%.

[0175] 2) Then, the coverage of each sample is normalized using the median method.

[0176] 3) Next, a logistic regression model was used to calculate the differential methylation levels in malignant and benign thyroid nodule samples. At the same time, covariate corrections were made for patient age and gender, and the calculated P-values ​​were corrected using the SLIM method.

[0177] 4) Finally, sites in malignant and benign thyroid nodule samples with an absolute difference in methylation level greater than a threshold (e.g., the threshold is set to 15%) and a significant difference (P value < 0.001) are defined as differentially methylated sites.

[0178] 5) The differentially methylated sites obtained through screening can be used as methylation markers. A total of 780 methylation sites were obtained. The 780 methylation sites and 273 genes associated with methylation sites are shown in Table 6.

[0179] Table 6

[0180]

[0181]

[0182]

[0183]

[0184]

[0185]

[0186]

[0187]

[0188]

[0189]

[0190] 2. Constructing a model for assessing the benign and malignant nature of thyroid nodules.

[0191] 1) A total of 49 samples were collected, including 30 malignant and 19 benign cases. 60% of the 49 samples were used as the training set, and the remaining 40% were used as the test set.

[0192] 2) Compare the difference in methylation levels of 780 methylation markers between malignant and benign thyroid nodule samples in the training set, and use the training set to train a random forest model (50000 trees, mtry = 5).

[0193] Thyroid nodule benign and malignant evaluation model construction: Form a data matrix (markers) of the methylation levels of methylation markers of 30 patients with malignant thyroid nodules and 19 patients with benign thyroid nodules, and store the classification information of 49 samples as a matrix (pheno) with the first column as the sample name, the second column as the classification information, the sample name of thyroid cancer as p.ca, and the sample name of benign sample as (n.normal).

[0194] The specific training process is as follows:

[0195] a) Open the R (V3.6.2) package, import the classification information matrix (pheno) and the corresponding methylation level matrix (markers) of 49 samples

[0196] pheno = read.delim(“storage path of classification information matrix”, sep = "\t", as.is = T, header = T, check.names = FALSE)

[0197] markers = read.delim(“storage path of methylation level matrix”, sep = "\t", as.is = T, header = T, check.names = FALSE)

[0198] b) Transpose the methylation level matrix to a matrix with sample name as row name and methylation marker name as column name

[0199] imput = t(markers)

[0200] c) Fill in the NA values with the nearest value

[0201] library(DMwR)

[0202] imputed = knnImputation(imput)

[0203] d) Set the modeling parameters

[0204] Library(randomForest)

[0205] mtry <- tuneRF(imputed, pheno, ntreeTry = 50000, mtryStart = 3,

[0206] stepFactor = 1, improve = 0.001, trace = TRUE, plot = TRUE)

[0207] e) Constructing random forest training model

[0208] mod_rf <- train(imputed, pheno, method = 'rf', mtry = mtry)

[0209] In the process of constructing the model, different codes can be assigned to the malignant type and the benign type, and a classification threshold is set, and the model is trained by R software and the random forest method, and finally the thyroid nodule benign and malignant evaluation model which can distinguish the benign and malignant samples based on the classification threshold is obtained.

[0210] In one specific embodiment, the classification of the malignant type is coded as 1, and the classification of the benign type is coded as 0. In the process of training the model by R software and the random forest, the classification threshold is set to 0.5 by default.

[0211] The constructed model finally distinguishes the benign and malignant samples with 0.5 as the classification threshold, and the finally trained model is the thyroid nodule benign and malignant evaluation model.

[0212] Then the classification effect of the random forest training model is verified by using the data of the test set. That is, the predicted value is obtained by calculating the data of the test set by the thyroid nodule benign and malignant evaluation model.

[0213] When the predicted value is greater than or equal to 0.5, it is determined to be malignant; when the predicted value of the model is less than 0.5, it is determined to be benign.

[0214] Figure 1 The important features of the thyroid nodule benign and malignant evaluation model constructed in this embodiment.

[0215] Figure 2 The ROC curve of the thyroid nodule benign and malignant evaluation model constructed in this embodiment.

[0216] From Figure 1 and Figure 2 It can be seen that the area under the ROC curve of the constructed thyroid nodule benign and malignant evaluation model is 100%, the sensitivity is 100%, the specificity is 91%, the negative predictive value is 100%, and the positive predictive value is 94%, indicating that the thyroid nodule benign and malignant evaluation model constructed by the present application has good prediction effect.

[0217] Example 3 Test of thyroid nodule benign and malignant evaluation model

[0218] In this embodiment, DNA methylation sequencing is performed on clinical samples, and the methylation levels of the methylation markers in the sequencing results are analyzed using the thyroid benign and malignant evaluation model constructed in Example 2. The steps include the following:

[0219] 3.1. Clinical samples

[0220] The thyroid nodule tissues of 14 patients with malignant thyroid nodules and 16 patients with benign thyroid nodules were extracted by the method of Example 1, and the extracted genomic DNA was sequenced by RRBS library, and the sequencing results were obtained.

[0221] 3.2. After the sequencing results obtained in step 3.1 of this embodiment are filtered and quality controlled, Bismark is used for reference genome alignment analysis, and then the methylation levels of the methylation markers in each clinical sample are calculated to form a sample-marker numerical matrix of 30 samples to be evaluated.

[0222] 3.3. The sample-marker numerical matrix of 30 samples to be evaluated obtained in step 3.2 of this embodiment is substituted into the thyroid benign and malignant evaluation model constructed in Example 2, and the probability of the clinical sample belonging to the malignant is calculated and output.

[0223] The specific process is as follows:

[0224] a) Open the R package and import the 30 sample-marker numerical matrix to be evaluated

[0225] valdata = read.delim("storage path of sample-marker matrix", sep = "\t", as.is = T,

[0226] row.names = 1, header = T, check.names = F)

[0227] b) Transpose the matrix to a matrix with sample names as row names and methylation marker names as column names

[0228] imput = t(valdata)

[0229] c) Fill in the NA values with the nearest value

[0230] Library(DMwR)

[0231] imputed = knnImputation(imput)

[0232] d) Import the thyroid benign and malignant evaluation model (RFmodelTC) constructed in Example 2 again

[0233] model = readRDS("RFmodelTCpath")

[0234] e) After that, the model evaluation is started and the predicted value of each sample is calculated, and according to the predicted value, the risk of each sample belonging to the thyroid nodule malignancy or the thyroid nodule benign is determined.

[0235] library(randomForest)

[0236] class = predict(model, imputed, type = "response")

[0237] probs = predict(model, imputed, type = "prob")

[0238] When the predicted value is greater than 0.5, it is determined that the thyroid nodule is malignant, otherwise it is determined that the thyroid nodule is benign.

[0239] The prediction results of the thyroid nodule benign and malignant evaluation model are compared with the clinical pathological results, and the results are shown in Table 7.

[0240] Table 7

[0241]

[0242]

[0243] From Table 7, it can be seen that the accuracy of the prediction result is 93% (28 / 30), the sensitivity is 93% (13 / 14), the specificity is 94% (16 / 17), the negative predictive value is 94%, and the positive predictive value is 93%, indicating that the thyroid nodule benign and malignant evaluation model has good classification effect.

[0244] The application obtains the methylation level by sequencing the DNA of the thyroid tissue, performing quality assessment on the sequencing results by FastQC, performing quality control filtering by fastp, performing reference genome alignment analysis by Bismark, and further applying a logistic regression model to calculate the differential methylation level in the malignant and benign thyroid nodule samples; the site with an absolute value of the methylation level difference greater than a threshold (for example, the threshold is set to 15%) and a significant difference (P value < 0.001) is defined as a differential methylation site, and 780 new methylation sites are screened, which are associated with 273 tumor occurrence and development related genes and involve Fc gamma R-mediated phagocytosis signal pathway; by detecting the methylation levels of multiple genes and multiple sites, the problems of low and unstable methylation signals of single gene or single site are overcome, thereby improving the sensitivity and specificity of detection. The methylation marker and the thyroid nodule benignity evaluation model provided by the application can be used for the auxiliary diagnosis of the benignity and malignancy of thyroid nodules; the accuracy of the model is 95%, the AUC is 100%, the sensitivity is 100%, and the specificity is 91%, which has a good prediction effect on the benignity and malignancy of thyroid nodules and can improve the accuracy of the clinical diagnosis of the benignity and malignancy of thyroid nodules. When the model of the application is used for clinical diagnosis, the accuracy of the prediction result is 93% (28 / 30), the sensitivity is 93% (13 / 14), the specificity is 94% (16 / 17), the negative predictive value is 94%, and the positive predictive value is 93%; the model not only has very high sensitivity, which helps to reduce missed diagnosis, is conducive to early treatment and improves the survival rate of patients, but also has a high negative predictive value, which can avoid unnecessary surgery for patients, reduce the social medical and economic burden, and improve the overall survival quality of thyroid nodule patients. In addition, the model for detecting the benignity and malignancy of thyroid provided by the application can directly give the prediction result of the benignity and malignancy by analyzing the methylation level of the methylation marker in the biological sample to be detected, thereby avoiding the subjectivity of result interpretation in ultrasonic imaging detection and cytopathology detection, improving the consistency and accuracy of result interpretation, and having clinical popularization and application value.

[0245] The above is only a preferred embodiment of the application, and is not a limitation on the form and essence of the application. It should be noted that those skilled in the art can make some improvements and supplements without departing from the method of the application, and these improvements and supplements should also be considered as the protection scope of the application. For those skilled in the art, some minor changes, modifications and equivalent changes made by using the disclosed technical content without departing from the spirit and scope of the application are equivalent embodiments of the application; at the same time, any equivalent change, modification and evolution of the above-mentioned embodiments according to the essential technology of the application are still within the scope of the technical solutions of the application.

Claims

1. Use of a substance for detecting 780 methylation sites in the preparation of a product for detecting the benign or malignant nature of a thyroid nodule.

2. Use according to claim 1, characterized in that, The product comprises a substance for detecting the methylation level of 780 methylation sites in Table 6 of the specification.

3. A product for detecting the benignity or malignancy of a thyroid nodule, characterized by, The product comprises a substance for detecting 780 methylation sites in Table 6 of the specification.

4. The product of claim 3, wherein, The product comprises at least one of a kit, a chip, a probe or a membrane strip.

5. A method for constructing a thyroid nodule benign and malignant evaluation model, characterized in that, The construction method comprises the following steps: 1) Constructing a sample data set based on the methylation level of 780 methylation sites in Table 6 of the specification, sample information and benign or malignant information in the biological sample of a subject, wherein the subject comprises a patient with a diagnosed benign thyroid nodule and a patient with a thyroid malignant nodule, assigning different classification codes to malignant samples and benign samples, and setting a classification threshold based on the classification codes; 2) Randomly dividing the sample data set into a test set and a training set, learning the data of the training set using a machine learning method, and obtaining the thyroid nodule benign or malignant evaluation model.

6. The construction method of claim 5, wherein, The biological sample is selected from at least one of plasma, serum, whole blood, urine, amniotic fluid and tissue; And / or, the machine learning method is selected from at least one of support vector machine, random forest algorithm and neural network algorithm; And / or, the division ratio of the test set and the training set is (5-7):(3-5); And / or, the thyroid nodule benign or malignant evaluation model is used to output a prediction value, when the prediction value is higher than the classification threshold, it is evaluated as a thyroid malignant nodule; when the prediction value is lower than the classification threshold, it is evaluated as a thyroid benign nodule; And / or, further comprising step 3) verifying the thyroid nodule benign or malignant evaluation model using the test set. 7.A device for constructing a thyroid nodule benign and malignant evaluation model, characterized in that, The device comprises: A data collection module for constructing a sample data set based on the methylation level of 780 methylation sites in Table 6 of the specification in the biological sample of a subject, wherein the subject comprises a patient with a diagnosed benign thyroid nodule and a patient with a thyroid malignant nodule, assigning different classification codes to malignant samples and benign samples, and setting a classification threshold based on the classification codes; a model construction module for randomly dividing the sample data set into a training set and a test set, and learning the training set using a machine learning method to obtain the thyroid nodule benign or malignant evaluation model.

8. The building apparatus of claim 7, wherein, The biological sample is selected from at least one of plasma, serum, whole blood, urine, amniotic fluid and tissue; And / or, the machine learning method is selected from at least one of support vector machine, random forest algorithm and neural network algorithm; And / or, the division ratio of the test set and the training set is (5-7):(3-5); And / or, the thyroid nodule benign or malignant evaluation model is used to output a prediction value, when the prediction value is higher than the classification threshold, it is evaluated as a thyroid malignant nodule; when the prediction value is lower than the classification threshold, it is evaluated as a thyroid benign nodule.

9. A device for assessing the benignity or malignancy of a thyroid nodule, characterized by, The device comprises: An input module for obtaining the detection amount data of 780 methylation sites in Table 6 of the specification in the biological sample of a subject to be tested; The benign and malignant evaluation module calculates the detection amount data of the input module by using the benign and malignant evaluation model of thyroid nodules obtained by the construction method of claim 5 or 6, obtains a prediction value, and compares the prediction value with a classification threshold to obtain a benign and malignant result of the thyroid nodules; The output module is configured to output the benign and malignant result of the thyroid nodules.

10. A method of detecting the risk of malignancy of a thyroid nodule, characterized by, The method comprises the following steps: S1. obtaining detection amount data of 780 methylation sites in a biological sample of a to-be-tested subject as shown in Table 6 in the specification; S2. calculating the detection amount data by using the benign and malignant evaluation model of thyroid nodules obtained by the construction method of claim 5 or 6, obtaining a prediction value, and comparing the prediction value with a classification threshold to obtain a benign and malignant result of the thyroid nodules; S3. outputting the benign and malignant result of the thyroid nodules according to the comparison result.

11. An apparatus comprising a processor and a memory, the memory to store a computer program, characterized in that, The processor is configured to execute a computer program stored in the memory, so that the device executes the construction method of claim 5 or 6 or the detection method of claim 10.

12. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the construction method of claim 5 or 6 or the detection method of claim 10.

Images