This is the first I have ever heard of a particular version of Ubuntu being best (or not best) for bioinformatics. My opinion is that all of them are the same, though more recent versions are preferred for security reasons. I can say this with some authority because I have computers that run Ubuntu 12.04, 16.04, 18.04 and 20.04. If you are installing the system from scratch, the only difference is that certain aptitude packages will not be available for the latest version (20.04) because of the normal lag. I don't rely on aptitude to install any bioinformatics packages, so this particular problem never comes up for me. It should make no difference what system version is there for installing python and R modules, though some older packages may only work with earlier python and R versions, and some may need the latest. In most cases that has nothing to do with the system version.
I do caution about upgrading the system to a newer version, as in my hands that almost always creates more work than it is worth. I don't update systems as a general rule, and the reason I have so many different versions is because these computers were put in production at different times - I always install the most up-to-date system available. About two months ago I deviated from this rule and upgraded a 18.04 system to 20.04. The upgrade went without any hitch and the computer is fully functional in general terms. As I ran various programs over the past few weeks, I have discovered that some python modules need to be reinstalled, or that some programs need to be re-compiled because the old libraries were removed. The same was true when about a year ago I was forced to upgrade two computers from 12.04 to 18.04 (security reasons), and had to do various tweeks to get everything to work as before. I know this may not be a popular opinion as many people upgrade as soon as the new version is available.
The choice of the distribution isn't that important. Choose any more or less current one. To install and manage your bioinformatics tools use conda with the bioconda channel activated. The first part of my tutorial might be helpful.
You should consider using LTS variant since they are guaranteed longer term support. If you are not a card carrying system administrator then staying with a newer release and keeping up with updates (program and security) would be a prudent choice.
Using conda to manage your software installations would again be advisable if you are not comfortable futzing with compiling things and finding libraries etc. This should in theory avoid problems with getting the programs to run as well.
Ultimately you need to conform to local IT security landscape. Check with your IT organization to stay compliant with any local policy (security or otherwise).
Are other users actively using this machine or are you the sole user? That makes a big difference in "administration" of a system and choices you make in practice.
We also had issues with Ubuntu 18.04. Ubuntu 16.04 is great but it reaches end of life in April. I've upgraded one machine from 18.04 to 20.04 and it seems to work fine. However, I use Ubuntu for desktops and laptops. For servers, I use CentOS although now that CentOS is not going to be the stable OS it used to be, I'll be looking for something else. Given that I already use Debian-based distros, I am thinking of moving my servers to Debian.