June 2023

Article by Riddhi Tatke In the realm of genomics, RNASeq has emerged as a powerful tool for unraveling the intricate workings of gene expression. By quantifying gene expression levels across different conditions or time points, researchers can gain valuable insights into biological processes, disease mechanisms, and even prognosis. Survival analysis, a statistical method originally developed in the field of clinical research, has now found its way into the analysis of RNASeq data, allowing scientists to explore the survival patterns of genes and their impact on various biological outcomes. In this blog post, we will delve into the fascinating world of survival analysis applied to RNASeq data. We will explore the fundamental concepts, methods, and applications of this approach, empowering you to unlock new layers of knowledge hidden within gene expression data. What is survival analysis? Survival analysis, also known as time-to-event analysis, is a statistical method used to analyze and predict the time until an event of interest occurs. In the biological context, survival analysis is particularly valuable for studying the time until an event (survival time), where the event can be death, relapse, disease recurrence, or any other significant outcome. These studies are used to understand the factors that influence the occurrence of these events and estimate the probability or risk of experiencing the event over time. By integrating survival analysis with RNASeq, researchers can identify genes whose expression patterns are associated with specific events or outcomes. There are various statistical methods used in survival analysis, such as the Kaplan-Meier estimator, log-rank test, Cox proportional hazards model, and parametric survival models. The Kaplan-Meier Estimator is a non-parametric method used to estimate survival probabilities and construct survival curves when analyzing time-to-event data. It is particularly useful when studying the survival times of individuals or groups in the presence of censored data. The Kaplan-Meier estimator allows for the comparison of survival probabilities between different groups or categories. The Log-Rank Test is a statistical hypothesis test commonly used to compare the survival experiences between two or more groups. It is applicable when the groups being compared are defined by categorical variables or when the variable of interest has a small range of values (such as high vs. low). The log-rank test evaluates whether there is a significant difference in survival times between these groups. Both the Kaplan-Meier estimator and the log-rank test are commonly employed in survival analysis, especially in situations where the variables under investigation are categorical or have a limited number of values. These methods can also be used with continuous variables (such as gene expression data), provided they are appropriately categorized or transformed. Cox Proportional Hazards Model is used for the analysis of continuous or continuous-covariate data. This model allows for the assessment of the influence of various variables, including continuous predictors, on survival outcomes while accounting for censoring. Commonly used terms related to survival analysis: Time to event: Time till an event (like death) occurs. Status: Whether the event occurred or not. Usually denoted by 1 (death occurred) or 0 (censored). Censoring: Survival studies typically have a specific duration of follow-up, during which the subjects are monitored for the occurrence of the event of interest (e.g., death, disease progression). Censoring occurs when the event of interest has not occurred for a particular subject by the end of the study or when they are no longer being actively monitored. There are 2 types of censoring: Right-censoring: This is the most common form of censoring in survival analysis. It happens when a subject has not experienced the event by the end of the study period. In this case, the survival time for that subject is unknown beyond the observed time point, and the data is considered right-censored. Left-censoring: This type of censoring occurs when the event of interest has occurred for a subject before the study started or before the subject entered the study. Left censoring is less common in survival analysis. Applications of Survival Analysis Survival analysis of RNASeq data has wide-ranging applications in prognostic biomarker discovery, disease classification, therapeutic target identification, elucidating biological mechanisms, predicting drug response, and facilitating personalized medicine. This powerful analytical approach paves the way for advancements in precision medicine and the development of tailored treatment strategies. Case Study A study by Zheng et al. aimed to analyze gene expression datasets related to gastric cancer and identify key genes associated with overall survival in gastric cancer. Five datasets from the Gene Expression Omnibus (GEO) were analyzed, and hub genes were identified using differential expression analysis and protein-protein interaction networks. The study utilized Kaplan-Meier survival curves to assess the correlation between the hub genes and the survival time of gastric cancer patients. Among the 59 hub genes identified, 21 showed no significant correlation with survival time, 31 had previously been reported to be associated with gastric cancer occurrence, and 6 genes were newly discovered to be associated with the prognosis of gastric cancer. These six genes, namely SERPINH1, NPY, PTGDR, GPER, ADHFE1, and AKR1C1, were found to be significantly associated with overall survival in gastric cancer, despite not having been previously reported in the context of gastric cancer. However, these genes have been reported in other studies related to different types of cancer. The study highlights the significance of survival analysis as a powerful tool for identifying genes directly associated with overall patient survival. In addition to potential biomarker identification, survival analysis plays a crucial role in evaluating treatment effectiveness and assessing the impact of different treatment strategies on survival outcomes. This aids in clinical decision-making and the development of evidence-based guidelines. Conclusion Survival analysis is a powerful approach that enables researchers to uncover meaningful associations between gene expression patterns and clinical outcomes. By exploring the relationship between genes and survival outcomes, we can gain valuable insights into disease prognosis, identify potential biomarkers, and pave the way for personalized treatment strategies. As technology evolves and our understanding of genomics deepens, survival analysis will undoubtedly continue to revolutionize the field of biomedical research, bringing us closer