-
Notifications
You must be signed in to change notification settings - Fork 47
Week 4
Task: Create a quality control module and routine to assist the user in determining whether the chosen sample is statistically compatible and comparable for their particular analysis.
Method: Develop a module that tests the data for normality and applies a log2 Transformation with the affy R Package and the justRMA() function which outputs log2-transformed expression values.
-
I have a method for editing the data table using shinyBS and a modal window. For right now I have a half working example in GEO-AWS_Test Repo. I'm going to migrate over the example to the actual GEO-AWS scripts by latest Wednesday of this week. Our BIG goal of survival analysis can't be completed if we don't have a way for the user to clean the clinical data table! I'm just having issue in regards to the observe and global variables of a data set. Hopefully this will be mediated soon enough.
-
The goal of this week is to provide a new module for the application that enables the user to decide if their chosen GEO Accession number sample is statistically "fair" to compare in the Differential Expression Analysis module or the eventual survival analysis.
-
The GEO2R tool doesn't have an elaborate method of competing this so we can use this R script as template to understanding and configuring our own module for this.
Work flow:
- display a boxplot of expression values
- provide options: auto-detect, Yes, No to determine if those values are on the log scale and what the user wants to do with them
- allow the user to proceed with the rest of the application once they have selected the options
I just implemented a way for the user to determine the groups to define for survival analysis!
-
Scratch that previous commit for the group selection boxes - I need to append checkboxes to each row in order to allow the user to perform survival analysis on all of the samples (and choose which rows to exclude). I also need to devise a method for multi-selecting rows based on using 'Shift'. Maybe I could have all of the rows pre-selected, similar to this example. I'm going to reverse that change commit.
-
My editable tables (on the cars/iris example) are not fully working and I think I found out why in this issue from Issue #20, and comment from Yihui Xie:
"datatable() knows nothing about the relationship between the new data and the old data"
- My current state of the example code has a work around to leave the find input box as the original and only change the replace box and then the table will update to the newest table.
- I have rudimentary quality control module to display the expression profiles and allow the user to select whether or not the samples look fair to compare and applying a log2 transformation to the data. From Wikipedia:
In statistics, data transformation is the application of a deterministic mathematical function to each point in a data set — that is, each data point zi is replaced with the transformed value yi = f(zi), where f is a function. Transforms are usually applied so that the data appear to more closely meet the assumptions of a statistical inference procedure that is to be applied, or to improve the interpretability or appearance of graphs.