If you create a list of genes longer than 23 in the RNA Seq section of the IVY Gap GBM section, only the first 23 are visible to take a screenshot.
Is it possible to create a list of more than 23 genes and then download or copy the whole image of genes within the heatmap ?
Hi @Stanley_Stylli,
Thank you for your interest in the IVY Gap project! There are two ways to capture the image of the heatmap: (1) taking a screenshot (which captures everything) or (2) right clicking on the image and clicking “save image as” (which captures just the heatmap image, and not the gene names, or other color bars). You can download up to 2000 genes using the “Download this data” link in the bottom right of the screen. You can also increase the gene list shown on the screen to more than 23 by zooming out in browser window (e.g., by holding “Ctrl” and spinning the scroll wheel on the mouse). I hope this helps!
HI Jeremy
I have an additional query regarding the heatmap.
When the data is downloaded using the download tab link, the resulting CSV file has both positive and negative numbers listed.
Am I correct in saying that these are the z-score values for each gene for a particular sample within a tumour region ?
If I want to compare gene expression values between two regions (eg) leading edge versus infiltrating tumour, can this be done with these normalized z-score values ?
Or is it best done with the gene level FPKM values ?
Thank you for your assistance.
It is greatly appreciated
Kind Regards
Stan
Hi Stan,
At the bottom right of the screen there is a color bar, which if you click on it will give you the following options:
By default, you are correct that the z-scores are shown, but you can change it to log2 intensity if you’d like. The “Download this data” button will download whatever data the heatmap is currently showing. As for which strategy is best, there are analysis methods that can deal with either value, but I prefer using FPKM. You can also do these analyses directly in the web app.
Best,
Jeremy
Hi Jeremy
Thanks for the tip with regards to changing the download data option from the color bar.
Just a couple of questions :
-
Are the FPKM values via the log2 intensity normalized ?
-
Also the link that you sent for carrying out the analyses via the web app, the differential search in this link shows 10965 genes in the list. Is it possible to carry out this search by inputting a smaller gene list ? (as your example is exactly what I would like to do – compare the Leading Edge versus the Infiltrating Tumour).
Thank you once again
Kind Regards
Stan
Hi Stan,
To address your question: (1) yes they are log2 normalized and (2) no you cannot perform the analysis on a subset of genes. All genes are input into the differential search and all genes with higher expression in target vs. contrast are shown (sorted by fold change). To find genes changing in the other direction, press this circle arrow button above the “Search” button to swap the target and contrast regions.
Best,
Jeremy