Glossary

A Glossary for terms used throughout PanHunter

Maintenance

If a project is in maintenance mode, users can’t access the apps from the Start Page. This can be useful for project initialisation or data integration.

Gene Ontology (GO) is a standard system for categorizing the function of genes in a comprehensive and meaningful biological framework. GO helps to understand the relationship between various genes and different biological pathways by assigning genes to terms. These standardized vocabulary are arranged in a hierarchy, with broad concepts at the highest and more specific ones at the lowest. There are three main GO branches:

Biological Process (BP): Correlates to the biological processes that the product of genes are involved in, such as immune responses, cell proliferation and division, signal transduction and so on.
Cellular Component (CC): Defines the specific locations within the intracellular environment in which the protein of interest is in its active form.
Molecular Function (MF): Corresponds to the biochemical activity of the product of a gene, such as catalytic, enzymatic, structural, transport, binding activities, etc.

P-Value

P-Value represents the probability of observing a extreme or more extreme result than what was actually observed, under the assumption that the null hypothesis is true. The null hypothesis, denoted H₀, is a statement of no effect/ differences between the reference and experimental samples in a biological trial. The decision regarding the acceptance or rejection of the H₀ is contigent upon the comparison between the p-value and the signficance level (α).

In the context of statistical hypothesis testing, α serves as a pre-determined treshold of probability utilized to assess whether the results of a study are statistically significant. It represents the highest level of probability at which the H₀ can be assumed true despite being false. The most commonly used α is set at 0.05 (5%), indicating that the researchers are willing to accept 5% risk of erroneously rejecting the null hypothesis. Nonetheless, a lower α (e.g. 0.01) implies researchers’ higher demand for stronger evidence to reject the H₀, while a higher α (e.g. 0.1) indicates researchers’ greater flexibility in accepting weaker evidence for supporting the H₀. A p-value less than the pre-determind α indicates that the likelihood of the observed result occuring by chance is very low. As a result, the H₀ is rejected, providing evidence to support the alternative hypothesis (Hₐ). Conversely, a p-value greater than α suggests that the observed result could have occured by chance, and the H₀ is not rejected. In such cases, there is not sufficient evidence to support the Hₐ.

In general, the p-value is the measure of strength of evidence against the null hypothesis.

Various equations can be employed to compute the p-value, which vary depending on the statistical analysis being conducted and the underlying assumptions about the data.

False Discovery Rate (FDR)

False Discovery Rate (FDR) is a statistical technique aimed at controlling the number of false positive results in multiple hypothesis testing by estimating the proportion of false positive results to the total number of results determined as statistically significant.