Edits for the results section

[jeizenga]

I have more comments in this section. Some of them are things that I was unclear on when I read it, but that also meant that I didn't quite know what to correct. I also made some more significant structural suggestions that I thought would be better to open up to discussion rather than to change myself in this PR.

I also expanded some of the figure captions. I think it's important to write figure captions so that they can be understood without reading the text, because people are going to skim the figures before they decide whether to read the paper. Unfortunately, that means including some repetitive information about the experimental set-ups.

[glennhickey]

Thanks @jeizenga!! Your PRs have been massively helpful.
@eldariont Can you address the remaining comments in the yeast section? I think they're mostly pretty simple. Some thoughts:

• Agreed that the mapping evaluation figures can move to the supplement (with something high level in the main text).
• A I mentioned in the comments, I think MAPQ & identity is stronger than just identity, but we do need to make that case.

[glennhickey]

On a side note: looking it up after seeing @jeizenga's edits, I just learned that while us Comonwealthers don't use Oxford commas, Americans ironically do.

[eldariont]

Yes, this is on my list for tomorrow. Thanks to both of you for the good comments!

[eldariont]

Okay, I moved the mappability figure and text into the supplement and only mention it shortly in the main text. I also swapped mapping quality and perc. identity in the figures and explained why both are important. I replaced Pearson correlation with Spearman rank correlation because our data is bimodal, not normal. I also explained that missing SNPs and Indels might be behind the poor performance of the VCF graph as discussed in #76.

[jeizenga]

I think it looks good