User:ElNando888/Static Cloud Labs/CL-VII.Frog Legs

The PDB database has been collecting entries for quite some time. The majority of the published 3D structures are proteins, but there has been a good number of RNA submissions as well. Unfortunately, many of those structures are duplicates and not every piece of RNA in nature is on file. Why talk about it? Well, it helps getting the right perspective when considering search results.

FRABASE has indexed most of the PDB database, and when a search for a 1-1-1-1 or 2-1-0-1 multiloops returns no hits, one may not conclude that the motif just doesn't exist in nature. But I believe it is a sign that these motifs are at least uncommon. And if so, then it is probably difficult to stabilize such a structure...

In the absence of natural examples, I pondered about the 2-1-0-1 multiloop. Given the tendency of RNA to form stacks in various situations (see the "loopy bulges"), it seems very clear that the adjacent stacks in this multiloop will tend to coaxially stack on each other. What does it mean spatially? Well, the free bases on the 1-nt-long segments should be quite close from each other, and the 2-nt-long segment should be facing them. With any luck, this could result in a 2-2 internal loop style motif, with a "twist" on one side.

O=O O=O

| | | |

O=O-X X O paired bases

| | x unpaired

O=O-X X = canonical bond

| | | | | and - backbone

O=O O=O

And then I recalled that the most successful 2-2 internal loop, in terms of SHAPE scoring, is the GG/GG one. So I asked Vinnie to generate a candidate with this constraint.

The 1-1-1-1 multiloop will certainly fold very differenly, but I felt curious about the all-G option as well.

User:ElNando888/Static Cloud Labs/CL-VII.Frog Legs

Links