Team:Worldshaper-Wuhan/Model
Modeling
References
Attachment
Modeling
It is well known that noncoding RNA including miRNA and long noncoding RNA (lncRNA) could regulate gene expression at transcriptional and post-transcriptional level [1]. The aberrantly expressed miRNA and lncRNA have been widely reported to be implicated in tumorigenesis and development of various cancers, including breast cancer [2-4]. Previous studies indicated mRNA, miRNA and lncRNA could achieve cross-talk between each other by forming a regulatory network [5]. Through miRNA response elements, lncRNAs could serve as ‘sponges’ for miRNAs, leading to changes in the miRNAs-regulated mRNA levels. Recently, emerging evidences have indicated that the lncRNA-miRNA-mRNA network might play vital roles in progression and metastasis of multiple cancers, including breast cancer [6].
Several lncRNAs, such as XIST, NEAT1, SNHG16, GAS2, and MALT-AS, have been reported to play an important role in tumorigenesis and development of breast cancer. However, different studies have different conclusions. Now, we wondered whether XIST, NEAT1, SNHG16, GAS5, or MALT-AS is involved in breast cancer.
But which one should we choose? As an indicator of breast cancer diagnosis，lncRNA expression level in cancer tumors should be significantly different from in normal cells. WE obtained our raw data about the expression levels of XIST, NEAT1, SNHG16, GAS5, and MALT-AS from TANRIC http://ibl.mdanderson.org/tanric/_design/basic/query.html. TANRIC is an open-access website that contains expression profiles of lncRNAs based on recent large-scale RNA-seq datasets. This website allows us to analyze our lncRNAs of interest.
Performing Statistic Analysis
We employed 2-sample t-test to investigate lncRNA exression levels in normal cells and tumor.
Assumption:
The null hypothesis (H0) is that the mean of lncRNAs expression level in normal cell is equal to the mean of lncRNAs expression in tumors (μ_1=μ_2)
The alternative hypothesis (H1) is that the mean of lncRNAs expression level in normal cell is not equal to the mean of lncRNAs expression in tumors. (μ_1≠μ_2).
Specify significance level: We followed the common convention and decided our significance level equal to 0.05 (α=.05).
Compute test statistic:
The t-test value of NEAT1 is 0.0216, and the t-critical table is checked. When the degree of freedom is equal to 1, the result is slightWhere X ̅1 represents the sample mean of the control group (normal cells), X ̅2 represents the sample mean of the experimental group (breast tumor cells), n1 represents the sample size of the control group, and n2 represents the sample size of the experimental group. S12 and S22 are two sample variances. This formula is one of the test principles, if the absolute value of the T-test calculated according to the sample data is greater than tα, n-1, the null hypothesis is rejected, and the population mean is considered to be unequal. Otherwise, the null hypothesis is not rejected.
Table 1. 2-samples t-test result
Based on statistics analysis result, we conclude that the null hypothesis of NEAT1, XIST, SNHG16, and MAL2-AS are rejected. For LncRNA NEAT1, XIST, SNHG16, and MAL2-AS, there are significant differences between the population means of the expression level in normal cells and tumors (t(940)=-2.025, P<0.05; t(940)=-7.930, P<0.05; t(940)=-3.500, P<0.05; t(940)=-5.617, P<0.05). However, we notice that the difference between the NEAT1 expression level in normal cells and tumors is small.
Fig.1 The expression of some lncRNAs in breast cancer by TANRIC.
Through above analysis, XIST and NEATs were expressed at lower levels in breast cancer tumors (Fig. 1). The expression level of SNHG16 and MALT-AS increased in breast tumors, compared to the control groups (Fig. 1). GAS5 expression level has no significant change in breast cancer tumors. LncRNA and microRNA are complementary and we have chosen microRNA with increased expression levels for later experiment. Therefore, both XIST and NEAT, with lower expression levels, met our requirement. Furthermore, because XIST has a larger expression level difference than that of NEAT, we choose XIST for our experiment.
Reference:
1. Esteller M. Non-coding RNAs in human disease. Nat. Rev. Genet. 12(12), 861–874 (2011).
2. Garzon R, Calin GA, Croce CM. MicroRNAs in cancer. Annu. Rev. Med. 60(1), 167–179 (2009).
3. Huarte M. The emerging role of lncRNAs in cancer. Nat. Med. 21(11), 1253 (2015).
4. Wang J, Ye C, Xiong H, Shen Y, Lu Y, Zhou J, Wang L. Dysregulation of long non-coding RNA in breast cancer: an overview of mechanism and clinical implication. Oncotarget. 2017 Jan 17;8(3):5508-5522.
5. Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP. A ceRNA hypothesis: the Rosetta stone of a hidden RNA language? Cell 146(3), 353–358 (2011).
6. Huang X, Xie X, Liu P et al. Adam12 and lnc015192 act as ceRNAs in breast cancer by regulating miR-34a. Oncogene 37(49),
The attachment:
