Project planning
I chose paper 4, which is about the whole genome assembly of the durian fruit. The aim of this project is to re-analyze the data from the paper and figure out which genes and pathways that are important for the different processes in the durian fruit, as well as compare transcriptomic data between different plant organs (fruit, aril, leaf, stem and root). This will be done in order to see whether the results from this project will differ from the results in the paper, as well as to gain knowledge about different bioinformatic tools and methods.
The analyses I will perform and the softwares I am going to use can be seen in the diagram below, figure 1. This image (figure 1) represents the workflow I am going to follow in order to finish this project.
Figure 1. Workflow of project. The analyses that will be done and the softwares that are going to be used can be displayed, as well as the input data and output data.
Some of these analyses that can be seen in the figure will take a longer time. These are Canu and BWA. Canu will take approximately 17 hours and BWA takes approximately 1 hour. If I have time, I will do additional analyses, such as for example transcriptome assembly and assembly evaluation with more than one method.
The project must be finished before may 18. Milestones for when certain steps in this project must be finished can be seen below in a bullet list:
- 6/4 - Preprocessing
- 8/4 - Compulsory computer lab
- 19/4 - Genome assembly
- 28/4 - Mapping
- 11/5 - Annotation
- 13/5 - Compulsory computer lab
- 17/5 - Analysis of result
- 18/5 - Hand in presentation and github
The data handled in this project is from an eukaryot and a whole genome assembly is to be made. I will obtain trancriptome data, PacBio reads and Illumina reads. PacBio reads are long-reads and Illumina reads are short-reads and are from the fruit stalk of the durian fruit Durio zibethinus Musang King. The transcriptome data is RNA-seq data and are from the stem, root, leaf and aril from the durio zibethinus Musang King. However, some of the arils are also from the durio zibethinus Monthong.
This will result in a big amount of data, which means a lot of space is needed in order to store the data. This may cause a problem if only storing in the home directory since I only got 32 Gb. However, if more space is needed, I will create a personal folder within the project directory and save the data there. I will also compress big data files in order to save space!
The metadata is going to be stored in an excel sheet and saved as a .tsv file or .csv file. I will have one sample per row and one variable per column, informative names and avoiding spaces when naming, as well as be consistent in order to avoid confusion.
The code and the data (such as figures, text, results) are going to be separated in my working directory. The code is going to be in a repository with version control (github) so I can track the coding process and be able to go back to a previous code that works if the current code has errors. I will have a readme file where I explain the workflow and in which order the codes are executed. The files will have informative and unique name so I do not get confused when I want to find a specific file. Folder and file names will start with a number, since then I can keep track on the order of when files/folders were created. Big data files will be compressed in order to save space.