No one thought of merely formatting the cells so as to not do this? Takes all of 3 seconds to do the entire spreadsheet, done. You can even do it after the fact 
If only it were that easy. That's the way it SHOULD work, and sometimes it's the way it does work. But there are also bugs in Excel that have persisted for decades, where data is converted upon importing it. The data is mangled and cannot be recovered. Nobody wants this behavior and there is no reason for it, but it's a very common problem.
In a sane spreadsheet, the source data and the human representation of the data should be two separate things. It should always be possible just to change the formatting if you don't like it. The spreadsheet should never over-write the source data unless the user explicitly changes it through the formula bar. Did you notice how many times I said "should"? In reality, Excel irreversibly mangles certain data in a way that you cannot recover it. This happens mostly when importing data and people have been complaining about it for literally decades. What the researchers are doing is probably the right thing. I take that back; the right thing is to not use Excel. So what they are doing is the pragmatic thing. The entire scientific community changing it's practices is more likely than Microsoft fixing bugs in Excel.
You can do just a single cell or a selection of cells or an entire column or roll or the entire spreadsheet at a time.
If only this were true. But I promise, interact with enough different types of data, and you will run into data-mangling, and you will find out there's nothing you can do about it, as many others have. As a fun exercise, I suggest importing some data with that has a minority of fields with leading zeros, and/or whitespace. After that, I suggest importing data that was exported using a different version of Excel, or the same version of Excel in a different region. Or, try importing some genetic data...
