I have some results tabulated in an Excel file, which consists of 42 fields and 73 records, each record comprised of 8 or 9 digit number (PDF attached). I want to represent it in a reader-friendly manner. Kindly guide me to represent the data in a better way.
If you want to give the values, well, then this table is ok (but I would place it in the supplement; I would also show them in rounded "millions" [x1E6] or show the logs with 2-3 significant digits).
If you want a graphical representation of all these values, then a heatmap is probably a good way to show it (possibly after hierarchical clustering).
Some analysis of the data may provide a sensible way of dimension reduction. A clustering may help. Or you do a principal component analysis (PCA) and show a biplot of the first components.
Thank you sir for your kind suggestion.
As I have mentioned in my question, these numbers are NCBI-GIs, which are actually gene IDs. I cannot round off them. Also heatmap representation will not be the right way to represent IDs. So, will representing such data in tabulated style be acceptable to the peers of any scientific journal ?
uh - oh... I did not read that XD
Gene IDs are no numeric values. They are like names. So a what-so-ever coding of supposedly numerical values is not sensible. Also any form of dimension reduction, clustering and PCA is nonsense for such data. Sorry.
There are 3066 IDs in the table, 42 of which are liste twice (=3024 unique IDs).
I do not see how you could condense this information.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication