BL5631 Practical Analysis of Genomic Data in R
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
BL5631 Practical Analysis of Genomic Data in R
CA1: Analysis of a multi-group transcriptomic experiment
1 Introduction to case scenario
We have chosen a number of interesting data sets that use Affymetrix microarrays and are slightly more than a two-class comparison problem. You may choose any one of these GEO series data for your study.
GEO ID |
Species |
Platform |
GSE153922 |
Human |
GPL16686 |
GSE71348 |
Human |
GPL16686 |
GSE50737 |
Human |
GPL16686 |
GSE51699 |
Human |
GPL16686 |
GSE56481 |
Human |
GPL16686 |
GSE57017 |
Human |
GPL16686 |
GSE124483 |
Mouse |
GPL21877 |
GSE124956 |
Mouse |
GPL21877 |
GSE81580 |
Mouse |
GPL21877 |
GSE136085 |
Mouse |
GPL21877 |
GSE147900 |
Mouse |
GPL21877 |
GSE140757 |
Mouse |
GPL21877 |
GSE126202 |
Mouse |
GPL21877 |
2 Assignment
You may work alone or in pairs to carry out a preliminary study to identify and interpret differ- entially expressed genes based on your choice of data set, working from the CEL file. deposited
on GEO.
Your analysis pipeline should be well-justified and the results reproducible (that is, anyone who replicates your study should be able to reproduce your findings).
Your report should be provided as follows:
• A PDF file, no more than 5 pages of text, plus figures and tables, in 12 point font.
• R code, as a github repository or uploaded to Canvas. The R code should be executable/.
2.1 Deadline
The assignment should be uploaded to Canvas before midnight, 20 September 2022.
3 Marking rubric
Your report will be graded based on the following criteria:
Design of study (20%) You should choose your dataset and set up your analysis to address the question at hand. A well designed study will indicate the choice of contrasts and the role of different contrasts in the interpretation of the study. Any limitations of the data set should be indicated. A report with excellent design will demonstrate a thorough understanding of the research question and provide justification for the choice of tools used in their analysis.
Documentation and code (20%) A well-documented report will include information neces- sary for reproducing the study, such as accession numbers and software packages used in the course of the analysis. Any code used should be provided as part of the report as a separate R file, or in a github repository.
Implementation and evaluation of results (20%) A report that demonstrates excellence in this area will provide insightful analysis of the results obtained and draw well-substantiated conclusions. Such a report will demonstrate sensible use of parameters (both default or otherwise), with justifications provided for the choice of parameters in the analysis.
Organisation (20%) A well-organised report will present the results in a logical and coherent prose, with the use of informative headings. Figures are well-used to highlight key results, and descriptive captions are provided to facilitate the interpretation of the figures.
Language and grammar (10%) An excellent report will be free of grammatical errors, have well-structured sentences and appropriate paragraphing.
References (10%) An excellent report will provide appropriate references where applicable, with a consistent citation style of your own choosing.
2022-09-17