> Rbjbj2uuC
^^8</4$ccccc_D$$$$$$$$$')R+$+$ccH$w w w Hcc$w $w w w cNR=bw #$0$w )q)w w b) w +$+$w $)^m: Responses to the Editor comments (Joram Posma)
Response: Thank you so much for your suggestions! We had considered your suggestions including the thoroughly proof-read the manuscript and correcting type-o's and addressing grammatical and spelling errors. Furthermore, we revised the manuscript as suggested by you and the reviewers, and provided a rebuttal letter. We look forward to hearing from you regarding our submission. We would be glad to respond to any further questions and comments that you and reviewers may have.
Responses to the reviewers comments:
Reviewer #1:
Experimental design
Even though the methodology and data are clearly described in the article, the whole text will benefit from proof reading to eliminate typos and various grammatical issues that appear across various sections including Figures. Nevertheless some presentations of methods I found a bit difficult to grasp, despite knowing something about their practical application.
Response: Thank you so much for your suggestion! We had considered your suggestion including the thoroughly proof-read the manuscript and correct type-o's and address grammatical and spelling errors. At the same time, we reedited the section of material and methods in this manuscript to make readers more understandable(line 102-146) .
Validity of the findings
1. More information should be provided on why the authors bounded the OS and DFS at 48 and 24 months respectively. The description: the curve tends to be gentle (line 138) is subjective. There is still at least a 10% decrease in both survival curves after those bounds (Figure 1 C-D).
Response: Thank you so much for your advice! We have considered your suggestion. The ROC curve is a necessary method to detect accuracy of this prediction model. we must define the outcome index of each patients at the first. However, there is no unified conclusion on the outcome index of the ROC curve in survival analysis. Some studies have used the median outcome index of the total survival time as the outcome index of the ROC curve(1), and others have used the outcome index of the follow-up endpoint (2). Based the two methods mentioned above, it could only assess the value of this model with corresponding time .As a result, we want to see the accuracy of this risk model in predicting the prognosis of every year. We draw the ROC for 1,2,3,4 year 0S and 1, 2 year DFS respectively in TCGA dataset, and their AUCs are 0.661,0.703,0.725,0.734 and 0.647,0.692(Supplementary Figure 1A-D and Figure 2E-F), respectively. Due to showing the highest predictive effect of this model in 4 and 2 year respectively .So, we identified the 4 year OS and 2 year DFS as the cuff-off time to draw the ROC, and we reedited the manuscript in line 180-186 to clear it.
1Zhu, X., Tian, X., Yu, C., Shen, C., Yan, T., Hong, J., Wang, Z., Fang, J., and Chen, H. 2016. A long non-coding RNA signature to improve prognosis prediction of gastric cancer. Molecular Cancer 15. 10.1186/s12943-016-0544-0,
2Qian, X., Nguyen, D.T., Dong, Y., Sinikovic, B., Kaufmann, A.M., Myers, J.N., Albers, A.E., and Graviss, E.A. 2019. Prognostic Score Predicts Survival in HPV-Negative Head and Neck Squamous Cell Cancer Patients. International Journal of Biological Sciences 15:1336-1344. 10.7150/ijbs.33329
2. Survival curves (figure1-4) should include a confidence interval. This is specifically important given that the size of the datasets is small.
Response: Thank you so much for your comments! As you mentioned, Survival curves (figure 1-4) should include a confidence interval, especially considering the size of datasets. So, we have added the 95% confidence interval of HR in Survival curves among the Figures.
3. It is not clear how the authors concluded that the small sample size in stage (1) patients is a limitation (line188) without carrying out a power analysis.
6. Table 3: number of patients doesnt match the text i.e. (line 189) says that a number of patients in stage I is 26, yet table 3 shows that stage I has 14 high and 41 low-risk patients. Please clarify
Response: Thank you so much for your comments! We rechecked the manuscript , finally, and found that the number of stage`!patients was 55( stage I has 14 high and 41 low-risk patients), definitely. So we have decided to delete the incorrect description about the small sample size in stage`!patients in manuscript, and the K-M survival curve including 55 patients in Figure 4A is still correct after we rechecked it.
4. Table 2: showing a representation of the workflow; somehow lacks the univariate cox proportional hazard regression analysis which isdescribed (line 129).
5. Table 2: layout should be modified as the directions of arrows follows a snake-like structure making it difficult to read. In addition, each rectangular box could be numbered to improve reading simplicity.
Response: Thank you so much for your comments! As you mentioned, layout should be modified as the directions of arrows follows a snake-like structure making it difficult to read, so we have added the univariate cox proportional hazard regression analysis to the table 2 and reedited the layout according to your advice to make it read easily(Table 2).
7. Age cut off for Old / Young is not specified (would be good to have clarification).
Response: Thank you so much for your comments! In fact, we used the age of 50 as the dividing line to divide the patients into the old or young. We have revised the manuscript (line 210)including the table 3.
8. P-value was adjusted (good to have clarification on how they were adjusted).
Response: To screen out differentially expressed lncRNAs. For example, if tens of thousands of gene expressions were measured in cancer patients and normal controls, to find out which genes may be different between the two groups of people, the student t test is generally enough. But after testing so many genes, false positives are inevitable(1-2). As a result, Bonferroni's correction has arised(1-2). Bonferroni correction adjusts the test level, resets the test level according to the number of indicators compared, and then draws a conclusion based on the P value. For example, the conventional student t test level is 0.05, as long as P is less than 0.05, it is considered to be statistically different. However, if adjusted using the Bonferroni's correction, you need to divide 0.05 by the number of genes compared. As a result, adjusted P value was widely used to reduce false positives in bioinformatic analysis such as screening out differentially expressed genes and enrichment pathways(3-4). In present article, Our setting test level is 3.5*10^(-6)(=0.05/14147) in fact (line 103-106).
1. Curtin, F., and Schulz, P. 1998. Multiple correlations and Bonferroni's correction. Biol Psychiatry 44:775-777. 10.1016/s0006-3223(98)00043-2
2. Sedgwick, P. 2014. Multiple hypothesis testing and Bonferroni's correction. BMJ 349:g6284. 10.1136/bmj.g6284
3.Deng, J., Xu, Y., and Wang, G. 2019. Identification of Potential Crucial Genes and Key Pathways in Breast Cancer Using Bioinformatic Analysis. Frontiers in Genetics 10. 10.3389/fgene.2019.00695
4.Zeng, J., Lu, W., Liang, L., Chen, G., Lan, H., Liang, X., and Zhu, X. 2019. Prognosis of clear cell renal cell carcinoma (ccRCC) based on a six-lncRNA-based risk score: an investigation based on RNA-sequencing data. Journal of Translational Medicine 17. 10.1186/s12967-019-2032-y
9. Figure 7: how the top 10 enrichment pathways were selected?
11. Could be explained how the Kyoto Encyclopaedia of Genes and Genomes was combined with the Web-based Gene set analysis toolkit to analyse the aberrantly activated signalling pathways.
Response: Thank you so much for your comments! To understand the potential functions of five lncRNAs, linear regression analysis was conducted to find the relationship between the five lncRNAs and the protein coding genes in TCGA. The screening criteria for the encoded protein genes was that these genes were positively associated with at least one lncRNA (Pearson coefficient > 0.8). After getting 3064 candidate genes, to understand the underlying enrichment pathway of those genes, we enter those genes into online websites Web-based Gene Set Analysis Toolkit (http://www.webgestalt.org/) according the manual (1), which has became one of the popular software tool in Bioinformatics including Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis(2-4). Web-based Gene Set Analysis Toolkit not only contain basic KEGG function, but also implement methods to call APIs provided by corresponding resources to view the pathway maps and highlight input genes in the maps(1). Therefore, this tool is a very practical and integrated tool for Bioinformatics(line 127-135). As other reports showed, Ranked by logP value (Q value), we selected the top 10 pathways for drawing bubble plot (line 222-228).
1.Wang, J., Duncan, D., Shi, Z., and Zhang, B. 2013. WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013. NUCLEIC ACIDS RESEARCH 41:W77-W83. 10.1093/nar/gkt439
2. Yang, M., Qin, X., Qin, G., and Zheng, X. 2019. The role of IRAK1 in breast cancer patients treated with neoadjuvant chemotherapy. Onco Targets Ther 12:2171-2180. 10.2147/OTT.S185662
3.Ding, L., Fan, L., Xu, X., Fu, J., and Xue, Y. 2019. Identification of core genes and pathways in type 2 diabetes mellitus by bioinformatics analysis. Molecular Medicine Reports 20:2597-2608. 10.3892/mmr.2019.10522
4.Xu, W., Li, J., Li, J., Yang, J.J., Wang, Q., Liu, B., and Qiu, W. 2019. An Investigation about Gene Modules Associated with hDPSC Differentiation for Adolescents. Stem Cells International 2019:8913287. 10.1155/2019/8913287
10. It might be more informative to include the clinical information for each cancer stage separately in table 3.
Response: Thank you so much for your comments! Because there is less clinical information about patients in TCGA data, the available clinical information we can extract is limited. As you can see, the clinical information of TCGA is basically shown in Table 3.
Also important to mention that the deficiencies in the statistical applications are only succinctly mentioned by the authors. In my view, it could have been identified in the article more meaningfully. Expanding it specifically to statistical methods as well as broadly discussing the shortcomings of data would make the paper more valuable.
Response: Thank you so much for your comments! It would be more valuable to expand the statistical methods broadly the shortcomings of data in section of discussion. We focused on this aspect at the end of the discussion (line 322-351).
Reviewer #2:
Basic reporting
1.LncRNAs and gene mentioned in the materials, methods and results section of the paper may express the same meaning. It is suggested to explain clearly and avoid confusion.
Response: Thank you so much for your comments! As you mentioned, we realized this problem. So we have reedited the whole manuscript to make it avoid confusion.
2.Please provide the lists of 278 lncRNAs and 38 shared genes in the supplementary materials.
Response: Thank you so much for your advice! We have updated such supplementary materials. You can check it in supplementary table 1. In the process of checking our work, we found that a total of 37 lncRNAs were included into Multivariate cox regression analysis. We feel really sorry for this mistake in our work.
3.Please correct the spelling mistakes such as time-dependent and forest plot.
Response: Thank you so much for your consideration and suggestions! As you mentioned, we realized this problem. We have completely revised the entire article to avoid such problems .
4.Figure 1B is not clear and please add the legend to explain it.
Response: Thank you so much for your suggestions! We have updated the clear heatmap and legend in Figure 1B. The legend annotated as follows: Heatmap of the five-lncRNA expression profile of the 414 patients in the TCGA dataset. Among five lncRNAs, MIR100HG and TRHDE-AS1 have a similar expression in 414 patients in the TCGA dataset, otherwise the other three lncRNAs do as well.
Experimental design
1.Please confirm the the cut-off value of high and low risk groups is the median or zero score in Figure 2C-D and 3C-D further.
Response: Thank you so much for your suggestions! To be precise, according to our calculation of each patient in the TCGA, the median of risk score is -0.001085 in fact. We all used the median of risk score to distinguish between high-risk group and low-risk group both in the training set and the validation set. We have corrected these descriptions in this manuscript (line 170-174).
2.How are 408 samples screened from the TCGA STAD database?
Response: Thank you so much for your consideration! We acquired a training dataset of gastric cancer samples from TCGA, comprised of 450 samples and 14147 LncRNAs (case: normal = 414:36). 450 samples were included to perform differential expression analysis. After that, excluding 6 cases with missing OS prognostic information, a total of 408 cases were recruited for further univariate Cox proportional hazards regression analysis and subsequent analysis in the training set. (line88-92).
3.DFS data of GSE62254 dataset is available from the supplementary materials of the article (Molecular analysis of gastric cancer identifies subtypes associated with distinct clinical outcomes), why is it not verified?
Response: Thank you so much for your comments! We rechecked the data downloaded from GEO. The DFS data of GSE62254 dataset is available. Then, we verified the risk model using the DFS data of GSE62254 dataset. The k-m survival curve, time-dependent ROC curve and scatter plot were shown in supplymentary figure 2. The results showed that this risk model have a predictive value in GSE62254 dataset (line192-201).
4.Please provide the details of the risk score model and the risk score of each patient to explain the range of risk score (-2.5-2.5) in the nomogram model.
Response: Thank you so much for your comments! An lncRNAs-based risk model was created from a linear combination of the expression levels of these lncRNAs, multiplied by the regression coefficients obtained from the multivariate Cox hazard analyses. Then, using the coefficients of five lncRNAs identified by multivariable Cox hazards analyses, we created a risk-score formula as follows: risk score = (0.249092 expression level of LINC00205) + (0.182045 expression level of TRHDE-AS1) + (0.271169 expression level of OVAAL) + ("0.20794 expression level of LINC00106) + (0.502539 expression level of MIR100HG). We have updated the risk score of each patient in TCGA in the supplymentary Table 2. The risk score in the TCGA ranged from -2.086959745 to 2.270305234 in fact. When we constructed the nomogram model including the risk score using nomogram R package, the range of risk score in this nomogram model outputted automatically transformed to (-2.5-2.5).
Validity of the findings
1.Please provide the multivariate cox analysis form of TCGA data.
Response: Thank you so much for your comments! We have provided the multivariate cox analysis form of TCGA data in supplymentary Table 3
2.The coefficients of 5 lncRNAs do not correspond to the HR values in Table 1.
Response: Thank you so much for your comments! After performing multivariate Cox hazards analyses, five independent prognostic lncRNAs were identified as independent factors. We rechecked our data, the result gained is correct in Table 1.
3.The sample sizes of the scatter plot (n = 300 and 140) are not consistent with that of data sets (n = 300 and 200) in the validation groups.
Response: Thank you so much for your comments! As you mentioned, we rechecked the data, and the patients in the validation group is 200. So, we corrected the scatter plot in figure 3D.
4.It is recommended to give the ROC analysis of1-4years cut-off OS and1-2years cut-off DFS.
Response: Thank you so much for your comments! As you mentioned, the ROC analysis of 4 years cut-off OS and 2 years cut-off DFS have been shown in figure 2E-F. We /89^_`ah ) 3 L M k m n
rcht5CJOJQJaJo(htCJOJQJaJo(h1oCJOJQJaJhyECJOJQJaJo(h*CJOJQJaJo(h1oh1oCJOJQJaJh1oCJOJQJaJo(h]S.h1oCJOJQJaJh1o5CJOJQJaJo(h1oh1o5CJOJQJaJh]S.h1o5CJOJQJaJ /
5
6
C
W
3
R#
2(
Px4 #\'*.25@9dh1$gd $a$gd~4dhgd)Hdhgd/'&$
2(
Px4 #\'*.25@9dh1$a$gd]S.dhgd]S.dhgd1o
5
6
B
C
V
W
Ư}m^m^mO@1hyEhyECJOJQJaJh]S.hzCJOJQJaJh/'>*CJOJQJaJo(hyE>*CJOJQJaJo(h/'h/'>*CJOJQJaJ2h/'h/'5>*B*CJKHOJQJaJo(ph/h/'h/'5>*B*CJKHOJQJaJph,hoE5>*B*CJKHOJQJaJo(ph3h]S.hoE5>*B*CJKHOJPJQJaJphh]S.hoECJOJQJaJh]S.hoE5CJOJQJaJ
1
ʶqY/h)Hh)H5>*B*CJKHOJQJaJphh]S.hoECJOJQJaJh]S.h6'CJOJQJaJh~4CJOJQJaJo(hkhCJOJQJaJo(h7$CJOJQJaJo(hjhCJOJQJaJo(hyEhyEhyEhyECJOJQJaJhvCJOJQJaJo(hyECJOJQJaJhyECJOJQJaJo(1
2
3
R[\v}~ξwj[TE9hKCJOJQJaJhKhKCJOJQJaJhKhKhyEhCJOJQJaJh~4CJOJQJaJo(h~4CJOJQJaJhCJOJQJaJo(h]S.hCJOJQJaJhhCJOJQJaJh]S.h5CJOJQJaJh)Hh)H>*CJOJQJaJ,h)H5>*B*CJKHOJQJaJo(ph3h]S.h)H5>*B*CJKHOJPJQJaJph5QR ";<NRS)+/8ITZjzʽ㰡ְֽ֒փփvjhdwCJOJQJaJhvCJOJQJaJo(hLhLCJOJQJaJhKhLCJOJQJaJh/h/CJOJQJaJh/CJOJQJaJo(h<CJOJQJaJo(hLCJOJQJaJhLCJOJQJaJo(hKhKCJOJQJaJhKCJOJQJaJo($jlx(fzȹsbsbUHh*CJOJQJaJo(hw[CJOJQJaJo(!h7$CJOJQJaJnHo(tH!hjhCJOJQJaJnHo(tHhyEhdwCJOJQJaJhdwCJOJQJaJo(hj9"CJOJQJaJhj9"CJOJQJaJo(hLhj9"CJOJQJaJhj9"hj9"CJOJQJaJhu^CJOJQJaJo(h<CJOJQJaJo(h<h<CJOJQJaJ|~wx)ŶŶŶxi\i\Ux\H\h_CJOJQJaJo(hu^hu^hu^CJOJQJaJo(h]S.hu^CJOJQJaJhu^hu^CJOJQJaJh]S.hu^5CJOJQJaJhu^>*CJOJQJaJo(hu^hu^>*CJOJQJaJh h CJOJQJaJh h CJOJQJaJo(h*h CJOJQJaJo(hCJOJQJaJo(h CJOJQJaJo(x#+\ ] !!!"&$&&%'')dhgd>BdhgdCdhgdu^#
2(
Px4 #\'*.25@9dh1$gd<)*5CV]`op"#ھ璂scTD5hu^h*CJOJQJaJh]S.h*5CJOJQJaJh>*CJOJQJaJo(hh>*CJOJQJaJh"G>*CJOJQJaJo(h"Gh"G>*CJOJQJaJh"GCJOJQJaJo(hV\hV\CJOJQJaJo(hV\hV\CJOJQJaJhV\CJOJQJaJo(hF3hF3CJOJQJaJhF3CJOJQJaJo(hu^CJOJQJaJo(hu^CJOJQJaJ&68:rHJbr|ȹլՒsլfZM>Mh&Oh&OCJOJQJaJh&OCJOJQJaJo(h
@.CJOJQJaJh
@.CJOJQJaJo(huJhuJCJOJQJaJhuJhuJ>*CJOJQJaJhuJCJOJQJaJo(hCCJOJQJaJo(hCCJOJQJaJo(hCh9QCJOJQJaJh9QCJOJQJaJo(hChCCJOJQJaJh*CJOJQJaJo(h]S.h*CJOJQJaJ.6J*+ʻ{n^O^@hQuS>*CJOJQJaJo(h
@.>*CJOJQJaJo(hQuShQuS>*CJOJQJaJhQuSCJOJQJaJo(h*CJOJQJaJo(hCJOJQJaJhCJOJQJaJo(hvCJOJQJaJo(h
@.CJOJQJaJo(hyEhI<6CJOJQJaJhI<6CJOJQJaJo(hCCJOJQJaJo(hCCJOJQJaJo(hChCCJOJQJaJ!)*= ! " : Q Z [ \ ] ĵĉĉyl_REhg5+CJOJQJaJo(hu^CJOJQJaJo(hjhCJOJQJaJo(h0N*CJOJQJaJo(hQuShQuSCJOJQJaJo(hQuShQuSCJOJQJaJh&Oh&OCJOJQJaJo(h&OCJOJQJaJo(h&Oh&OCJOJQJaJhQuSCJOJQJaJo(h]S.hQuSCJOJQJaJh"G5CJOJQJaJo(h]S.h"G5CJOJQJaJ] !!3!C!D!E!F!Y!ymy^y^y^yQEQhECJOJQJaJhECJOJQJaJo(hI<6hI<6CJOJQJaJhI<6CJOJQJaJhI<6CJOJQJaJo(hg5+CJOJQJaJo(h]S.hg5+CJOJQJaJhg5+5CJOJQJaJo(h]S.hg5+5CJOJQJaJ"hg5+hg5+>*CJOJQJaJo(hg5+>*CJOJQJaJo(hg5+>*CJOJQJaJhg5+hg5+>*CJOJQJaJY!c!d!k!m!n!!!!!!!!!!!!!"""佰zmaTE6hAhACJOJQJaJh>Bh>BCJOJQJaJh>BCJOJQJaJo(hACJOJQJaJhACJOJQJaJo(hAhAh]S.hC5CJOJQJaJh >*CJOJQJaJo(h}eh}e>*CJOJQJaJh CJOJQJaJo(hCJOJQJaJo(h7$CJOJQJaJo(hSCJOJQJaJo(hCCJOJQJaJo(hyEhECJOJQJaJ"".#3#5#B#Y#Z#k#v##/$9$C$$$%%%%%,%<%?%E%U%V%Z%g%h%o%%%%%%%%%Ǹ֩uhh5CJOJQJaJo(h;CJOJQJaJh>Bh;CJOJQJaJh;CJOJQJaJo(h>BCJOJQJaJhAh>BCJOJQJaJhAhw[CJOJQJaJhw[hw[CJOJQJaJhw[CJOJQJaJo(hAhACJOJQJaJh>BCJOJQJaJo(&%%%%%%%% &&&&&& &#&$&&&&&&&$'%'&'''ؾ籤症xhaxQDh>BCJOJQJaJo(hAhw[CJOJQJaJo(huJhuJhw[h>BCJOJQJaJo(huJhuJCJOJQJaJhw[hw[CJOJQJaJo(hw[CJOJQJaJo(h7$CJOJQJaJo(hSCJOJQJaJo(hLCJOJQJaJo(huCJOJQJaJo(h5h5CJOJQJaJh5CJOJQJaJo(h5CJOJQJaJ''''')))C)D)))* *$*,*-*.*/*A*ʽo`QDQD:/hKCJOJQJo(hKCJOJQJhKCJOJQJaJo(h]S.hKCJOJQJaJhK5CJOJQJaJo(h]S.hK5CJOJQJaJhOKa>*CJOJQJaJo(hOKahOKa>*CJOJQJaJh1s>*CJOJQJaJo(h1sh1s>*CJOJQJaJhGCJOJQJaJo(hCCJOJQJaJo(h>BCJOJQJaJo(hw[CJOJQJaJo(h>Bh>BCJOJQJaJ))D))..f/ 0011O2T3U345555d6dhgd]S.&$
2(
Px4 #\'*.25@9dh1$a$gd]S.dhgdu^dhgdGdhgdK dhgd8;dhgd>BA*R*S*c*d*e*f**********#+$+\+]+}+~+++++++++++++++++,4,O,^,d,e,̿杒{nhEhECJOJQJhQv-hKCJOJQJhECJOJQJhECJOJQJo(hKhKCJOJQJhSCJOJQJo(hKCJOJQJh$hKCJOJQJhKhKCJOJQJh1
*hKCJOJQJhKCJOJQJo(h1
*hKCJOJQJo(+e,f,g,h,j,,,,,,,--$-%-A-I-S-]-^-r-s------..).7.8.9.:.жЯuuunuucuYhWCJOJQJh`7CJOJQJo(h;(Qh;(Qh;(Qh;(QCJOJQJh]"CJOJQJo(h;(QCJOJQJo(h;(QCJOJQJhKh;(QCJOJQJh2h2hKhKCJOJQJhhCJOJQJhKCJOJQJo(hQv-hKCJOJQJo(hOKaCJOJQJo(hECJOJQJo(!:.R.[.\.].b.g.q.............e/h/0"00000111!2ܼǦqahGhG>*CJOJQJaJhKh1sCJOJQJaJo(hkCJOJQJhOKahOKaCJOJQJhOKahOKaCJOJQJo(hKCJOJQJo(h7$CJOJQJo(h8;CJOJQJo(h2CJOJQJo(h2CJOJQJh2h2CJOJQJhSCJOJQJo(hWCJOJQJo(!2"2N2O2X2Y2t2|2}2~2S3T3U3C4b4c444Ͽxk[N?[-"hltchltc>*CJOJQJaJo(hlT>*CJOJQJaJo(hlT>*CJOJQJaJhlThlT>*CJOJQJaJhuJCJOJQJaJo(hkrHhkrHCJOJQJaJhkrHCJOJQJaJo(hGCJOJQJaJo(h]S.hGCJOJQJaJhG5CJOJQJaJo(h]S.hG5CJOJQJaJhG>*CJOJQJaJo(hGhG>*CJOJQJaJ"hGhG>*CJOJQJaJo(4444444444O5Q5R5T5U5555555555ѷrererVD"h7$hltc>*CJOJQJaJo(h}e5CJOJQJaJo(h7$CJOJQJaJo(hDCJOJQJaJo(h`&CJOJQJaJo(hNhNCJOJQJaJo(hNhNCJOJQJaJhNCJOJQJaJhNCJOJQJaJo(hltcCJOJQJaJo(h]S.hltcCJOJQJaJhltchltcCJOJQJaJh]S.h}e5CJOJQJaJ55555c6d6m666666666ζ}m^Q^QD8D)h`&h`&CJOJQJaJh`&CJOJQJaJh`&CJOJQJaJo(hfdCJOJQJaJo(h]S.hoECJOJQJaJh]S.hoE5CJOJQJaJhfd>*CJOJQJaJo(h`&h`&>*CJOJQJaJ2h`&h`&5>*B*CJKHOJQJaJo(ph/h`&h`&5>*B*CJKHOJQJaJph,hoE5>*B*CJKHOJQJaJo(ph3h]S.hoE5>*B*CJKHOJPJQJaJph66777c7d7m7777777777+828F8K8R8j8k8l8m8vjv[v[v[vNBhWCJOJQJaJhWCJOJQJaJo(hhhhCJOJQJaJhhCJOJQJaJhhCJOJQJaJo(h`&h`&CJOJQJaJh]S.hoECJOJQJaJh]S.hoE5CJOJQJaJhb:>*CJOJQJaJo(h`&h`&>*CJOJQJaJ*h`&B*CJKHOJQJ^JaJo(phHHHh`&CJOJQJaJh`&CJOJQJaJo(d677d7888999f;g;{;;}===??@!B"BBdhgdjx&$
2(
Px4 #\'*.25@9dh1$a$gdjxdhgdWdhgd]S.m88888889192939D9E9_9999999ƶ{k\OB2hWhW>*CJOJQJaJhjCJOJQJaJo(h9 [CJOJQJaJo(hWhWCJOJQJaJhWhWCJOJQJaJo(h]S.hyCJOJQJaJh4hCJOJQJaJo(h]S.hCJOJQJaJh]S.h5CJOJQJaJh`&h>*CJOJQJaJh`&h`&>*CJOJQJaJh`&>*CJOJQJaJo(h`&CJOJQJaJo(hWCJOJQJaJo(999::::U:V:b:i:v:w::::::::::::;;ҸũthZOJZO h[6o(h[6CJOJQJo(hoh[6CJOJQJo(h[6CJOJQJaJh`4CJOJQJaJh67h[6CJOJQJaJh@syCJOJQJaJo(h[6CJOJQJaJo(h[6h[6CJOJQJaJh0CJOJQJaJo(hWCJOJQJaJo(h]S.hWCJOJQJaJh]S.hW5CJOJQJaJhW>*CJOJQJaJo(;/;F;Y;e;f;g;z;{;;;< <+<,<-<˵vdTE8E8hjxCJOJQJaJo(h]S.hjxCJOJQJaJh]S.hjx5CJOJQJaJ"hjxhjx>*CJOJQJaJo(hjxhjx>*CJOJQJaJ,hjx5>*B*CJKHOJQJaJo(ph/hjxhjx5>*B*CJKHOJQJaJph*hjxB*CJKHOJQJ^JaJo(ph4hWCJOJQJaJo(hpCCJOJQJaJo(h[6CJOJQJaJo(h`4CJOJQJaJo(-<:<<<<<<<<=== =3=4=5=Y=l=t=v=y=z={=|=}=====зqbRh]S.h n5CJOJQJaJh n>*CJOJQJaJo(h nh n>*CJOJQJaJhWCJOJQJaJo(h7$CJOJQJaJo(hfCJOJQJaJo(h nh nCJOJQJaJh nCJOJQJaJo(h&CJOJQJaJh&CJOJQJaJo(h)24h)24h)24CJOJQJaJo(h)24h)24CJOJQJaJ======???????@@@@@@@@ʿtdUH;U;h@CJOJQJaJo(hRCJOJQJaJo(h]S.hRCJOJQJaJh]S.hR5CJOJQJaJhR>*CJOJQJaJo(hRhR>*CJOJQJaJhRCJOJQJo(h nCJOJQJo(h]CJOJQJo(hnCJOJQJo(hpCCJOJQJo(hf
hf
CJOJQJh]CJOJQJaJo(h nCJOJQJaJo(h]S.h nCJOJQJaJ@@@@@AA
AAAA.A8ATAVAWA]AlAtA}AAAABBBBBB B!Bʾʱʤʱ|ob|Rh@hH1CJOJQJaJo(hf
CJOJQJaJo(hSRCJOJQJaJo(hpCCJOJQJaJo(h@hTUCJOJQJaJhTUCJOJQJaJhTUCJOJQJaJo(hU"CJOJQJaJo(h@CJOJQJaJh@CJOJQJaJo(h@h@CJOJQJaJhmzCJOJQJaJhmzCJOJQJaJo(!B"BBBBBBBBBCCCCCCCCCCCCCD,D0D4DDDôxj]Pj]x]x]x]FxFxh"qCJOJQJhch"qCJOJQJhoh"qCJOJQJhoh"qCJOJQJo(h"qCJOJQJnHo(tHh"qh"qCJOJQJaJ!h"qCJOJQJaJnHo(tHhXCJOJQJaJo(h]S.hXCJOJQJaJh]S.hX5CJOJQJaJhTU>*CJOJQJaJo(hTUhTU>*CJOJQJaJhH1CJOJQJaJo(DDFDJDnDpDDDDDDEEEXEhElEnEEEEEEF
F+F,F-FFFFFFFFFFFXG佰zi!h"qCJOJQJaJnHo(tHh"qCJOJQJaJh"qCJOJQJaJo(hXhXCJOJQJaJh$YAh"qCJOJQJo(hXCJOJQJaJo(hXCJOJQJaJhch"qCJOJQJhoh"qCJOJQJo(h"qCJOJQJnHo(tHhoh"qCJOJQJ%XGaGbGoG|GGGGGGGGGrZ@0hThT>*CJOJQJaJ2hThT5>*B*CJKHOJQJaJo(ph/hThT5>*B*CJKHOJQJaJph$hT>*CJOJQJaJnHo(tH$hX>*CJOJQJaJnHo(tH$h"q>*CJOJQJaJnHo(tHhTUh"q>*CJOJQJaJ!hDLCJOJQJaJnHo(tH$h"qh"qCJOJQJaJnHtH!h"qCJOJQJaJnHo(tHh"qCJOJQJaJnHtHBGGGGpHqHHII?JJJJJ[KLdhgd!dhgd]S.&$
2(
Px4 #\'*.25@9dh1$a$gdT$WDd`a$gdTm$$WDd`a$gd"qm$GGGHHHHHH%H&HVHpHqHHHHHHHHHHŹŪŪŝ~n_PPDh?7CJOJQJaJh]S.hL(CJOJQJaJhL(hL(CJOJQJaJh]S.hL(5CJOJQJaJhL(>*CJOJQJaJo(hL(hL(>*CJOJQJaJhL(CJOJQJaJo(hThTCJOJQJaJhTCJOJQJaJhTCJOJQJaJo(h]S.hTCJOJQJaJh]S.hT5CJOJQJaJh]q>*CJOJQJaJo(HI
III!I4I>I?IFIoIpIrI{IIII>J?JHJIJdJlJmJnJڸ{l]P]ChICJOJQJaJo(hDLCJOJQJaJo(h]S.hDLCJOJQJaJhL(hDLCJOJQJaJh]S.hDL5CJOJQJaJ"hL(hDL>*CJOJQJaJo(hDLhDL>*CJOJQJaJhDLCJOJQJo(hL(CJOJQJo(hQih?7CJOJQJh?7CJOJQJh?7CJOJQJo(hoh?7CJOJQJh?7CJOJQJaJo(nJoJ}J~JJJJJJJJJJJJJZK[KdKeK;Ͳͥ͋͘|l]M>hL(h!CJOJQJaJh]S.h!5CJOJQJaJhlKI>*CJOJQJaJo(hlKIhlKI>*CJOJQJaJh!>*CJOJQJaJo(hL(CJOJQJaJo(h-mCJOJQJaJo(hlKICJOJQJaJo(hx5CJOJQJaJhx5hx5CJOJQJaJhx5CJOJQJaJo(h!CJOJQJaJo(hICJOJQJaJo(hICJOJQJaJeKKKKKKKKKKKKKKKKK'(?QŠ!˼˼˧˼˰˚˧˼ˊthth CJOJQJaJhHih CJOJQJaJhHih h h CJOJQJaJo(h CJOJQJaJo(hHihHiUhHiCJOJQJaJhHihHiCJOJQJaJhHiCJOJQJaJo(h!CJOJQJaJh!CJOJQJaJo(h]S.h!CJOJQJaJ(also have showed the ROC analysis of 1-3 years cut-off OS and 1 years cut-off DFS in the TCGA dataset in supplymentary figure 1. Furthermore, we have showed the ROC analysis of 1-3 years cut-off OS in two validation datasets in supplymentary figure 3. The ROC analysis of 4 year cut-off OS in two validation datasets was shown in figure 3E-F. The ROC analysis of 1-2 years cut-off DFS in GSE 62254 was shown in supplymentary figure 2C-D. The parameter summary table was shown in Table 4.
5.In Figure 7A, the absolute value of Pearson correlation coefficient should not be greater than 1
Response: Thank you so much for your comments! As you mentioned, we realized that Pearson correlation coefficient should not be greater than 1. Based on this, we rechecked our data obtained, and the screening criteria for the protein-coding genes was that these genes were positively associated with at least one lncRNA (Pearson coefficient > 0.4). After identifying the candidate genes, aberrantly activated signaling pathways were screened out using the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis. We have corrected the data, the plots (figure 7A-B) and the descriptions (line 220-228).
![knst}ѡKLUVqyz{|whYYM@@hZ;$CJOJQJaJo(h!CJOJQJaJh]S.h!CJOJQJaJhL(h!CJOJQJaJh]S.h!5CJOJQJaJ"hlKIhlKI>*CJOJQJaJo(hlKIhlKI>*CJOJQJaJ"h!h!>*CJOJQJaJo(h!CJOJQJaJo(h h CJOJQJaJh CJOJQJaJhHih CJOJQJaJh CJOJQJaJo(ڢܢݢ
5<BCKLУ
7VXY[r˳ynynyhXCJOJQJo(hSRCJOJQJo(hA^CJOJQJo(hKhnCJOJQJhnCJOJQJh$hnCJOJQJhnCJOJQJo(h
CJOJQJaJo(hnCJOJQJaJo(hZ;$CJOJQJaJhZ;$CJOJQJaJo(hZ;$hZ;$CJOJQJaJ)huejhueUhThL(>*CJOJQJaJhZ;$h!CJOJQJaJo(gdk2dhgd]S.0182P. A!"#$%Sj666666666vvvvvvvvv666666>6666666666666666666666666666666666666666666666666hH66666666666666666666666666666666666666666666666666666666666666666p62&6FVfv2(&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv&6FVfv8XV~ OJPJQJ_HmH nHsH tHJ`JoEcke$1$a$ CJKH_HaJmH nHsH tH$A $
؞k=W[SOBiB
0nfh*B*ph.A.`>-baf-cellVoRV"qnormaldm$$CJOJQJ^J_HaJmH sH tH L bL~4e$1$a$ CJKH_HaJmH nHsH tHPK![Content_Types].xmlj0Eжr(Iw},-j4 wP-t#bΙ{UTU^hd}㨫)*1P' ^W0)T9<l#$yi};~@(Hu*Dנz/0ǰ$X3aZ,D0j~3߶b~i>3\`?/[G\!-Rk.sԻ..a濭?PK!֧6_rels/.relsj0}Q%v/C/}(h"O
= C?hv=Ʌ%[xp{۵_Pѣ<1H0ORBdJE4b$q_6LR7`0̞O,En7Lib/SeеPK!kytheme/theme/themeManager.xmlM
@}w7c(EbˮCAǠҟ7՛K
Y,
e.|,H,lxɴIsQ}#Ր ֵ+!,^$j=GW)E+&
8PK!mQtheme/theme/theme1.xmlYMoE#F{oc'vGuرhF[x=ޝzvg53NjHHzZ $ʯI)*E_N&IA!>Ǽ3xN>UWL0Č߫߿z}rx{h9T8T/χ?xY'~3S~O_|w)ӈHt=a+d NG1-RlƁ1RJwT蠯M1K"o
he˓ێ½PL-|%EhY7'qP.\L=dqķ3Iofi.ñBRb-JP_pG
ݢiKtdӌhFio7;7Q2ȾD>a/Q>XW
˔M_uHq)h+:Viw4rBqϫ"r!2laCbU|!8^8>ܠ,AбVt~ls14_=,ɬoTV G"ѦbH'.4xw-ϷE|F;v`b3"G'e\fHO'$i_wp >*8i&LY%\,xҕ=r6f 3Tʅ):ªZKLs&CMܛ0 [˫p@ע`j۽7YHxHicT5Ar@HZAZC}i' RQ\m,zo,gQu{Y\,Nר/=L_.̇Y7C6-fSh62"5NHT[X65̫4X%Yֳ2fkhi~tCKF#b+w1m|4`!:U!p5a:~{4mmioήc8mDJpSǹ橠V1?#Si?3E'pS2W`qB]( 08w
nA[s)k8= Aa?R dڒɾcUӽ˲d)#Queb}C!&i0>4S7{zF}4sKvUKoȳh~1jYUVH5U8Vk;֜L9ŰD ?*|FL
"A3>g]dm2iVֵ褽mg
C
1
)] Y!"%''A*e,:.!2456m89;-<=@!BDDXGGHnJeK!&()*+,-/012345679:;<=>?ABCDEFGHIKLMNSTU)d6B'.8@JVT #@H 0(
0(
B
S ? OLE_LINK8_Hlk35302031
OLE_LINK81
OLE_LINK82 OLE_LINK2 OLE_LINK5
OLE_LINK62
OLE_LINK63W**!9!9CW**)9)9C"'(-[^
")wBN^dkuYfbn{13fk|iv\c]c]"a"""##### #0#F#R#e#######$$5$7$I$L$$$$$6%9%x%}%%%%%(+/++,2,S,V,5.<.w.~.....R/Y///X2_263@3556666R7Y777j9w999::I:Q:::8;;;Z;g;;;?<F<g?t???@@BBCCCCCCCCCCC
=|~mw$/^eHz%+L "*""" $&$W$$$$%%U''))++,,----//7/{//113333g55"6*6::q;v;<<==@@BDBVCCCCCCCCCCC333333333333333333333333333333333333333333333E%0#e##$%%)),*c*g/z/51133==@@VCCCCCCCCCCC^q ijg1Ws
x}KK
x})Hkhjxv f
-Q$ E/'GTlTBQ j9"Z;$%%-!&L(dx(T)0N*g5+n-g8.
@.]S.H1k2F3XD3)24`4x5I<6[6?7`7U9b:8;U<)>CBfBpCyE"GW+HkrHlKIuJKKN&O;(QSRpSQuSSaUWXZ9 [V\A^u^7`K`OKa/b,^bltcfd>f"Ag4hHi-m=m!?mTn1o]q1sanstuwrwx@syjB{s{]
*<b!TUmzSL II/?Cjp 2pE
bf6'W nCFhAXk>BueNeRn@f/;E0J5C!X$s&PjhU"DLv
gw[}e"qLn]"_(T~4`&9Q7$_yzDKy:oEdw-\+CC@CC0CC|
89>C@,@<@D@@@Unknown G*Ax Times New Roman5Symbol3.*Cx Arial;([SOSimSun3*AxTimes7.[ @Verdana;"Helvetica7.@CalibriA$BCambria Math qh3gHGc
9 "z
9 "z?!%),.:;>?]} & 0 2 3 : !6"000 00
0000006:>@DZ\^ =@\]^$([{ 0
0000000Y[];[2CC OQHX $PoE2!xx
AdministratorJUJUMAOOh+'0\
$
0<DLTAdministratorNormalJUJUMAO99Microsoft Office Word@ne)@dS@}=
9՜.+,0
X`pxйz"C
!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWYZ[\]^_abcdefghijklmnopqrstvwxyz{|~Root Entry Fp1`=Data
X1Table`)WordDocument
2SummaryInformation(uDocumentSummaryInformation8}MsoDataStore=NR=YRYEOU3HSNOA==2=NR=Item
PropertiesUCompObj
u
F#Microsoft Office Word 97-2003 ĵ
MSWordDocWord.Document.89q