Probing the “Dark Matter” of Protein Fold Space

Abstract

We used a protein structure prediction method to generate a variety of folds as alpha-carbon models with realistic secondary structures and good hydrophobic packing. The prediction method used only idealized constructs that are not based on known protein structures or fragments of them, producing an unbiased distribution. Model and native fold comparison used a topology-based method as superposition can only be relied on in similar structures. When all the models were compared to a nonredundant set of all known structures, only one-in-ten were found to have a match. This large excess of novel folds was associated with each protein probe and if true in general, implies that the space of possible folds is larger than the space of realized folds, in much the same way that sequence-space is larger than fold-space. The large excess of novel folds exhibited no unusual properties and has been likened to cosmological dark matter.

Biophysics of protein evolution and evolutionary protein biophysics

“…Third, many of the lowest-energy 2D HP structures are highly compact but not maximally compact, as for real globular proteins [158,299], and a significant fraction of compact structures are not encodable by any HP sequence as its unique native structure [158,162]. The latter observation may bear on the question of whether the currently known set of globular protein folds is nearly complete in its coverage of all physically possible compact folds, as discussed in §3.1 [286][287][288][289][290], but one has to also keep in mind that the HP model interaction potential is less heterogeneous, and thus entails fewer encodable structures, than model potentials that contain repulsive interactions or otherwise more heterogeneous interactions [158,322,323]. Fourth, the 2D HP lattice model provides sequences that act like evolutionary bridges [161, 178,204,239,299] (see §3.5), encode for autonomous folding units [292,316,324], and exhibit homology-like behaviours [295], all similar to properties observed in real proteins.…”

Unexpected features of the dark proteome

“…4B and Fig. S8) with high occurrence of disulfide bonds is a signature, suggesting that many dark proteins are newly evolved folds (35) exploring the dark matter of protein folding space (12).…”

“…The analogy to dark matter has inspired surveys of other "unknown" properties of proteins; for example, Levitt (6) examined "orphan" protein sequences that do not match to known sequence profiles, which he termed the "dark matter of the protein universe," and Taylor et al (12) investigated the "dark matter of protein fold space" (i.e., theoretically plausible folds that have not been observed in native proteins). The same analogy has been used in studies of so-called "junk DNA" (13), which revealed a "hidden layer" of noncoding RNAs (14).…”

Are viruses a source of new protein folds for organisms? – Virosphere structure space and evolution

“…The virosphere structure space may contain structurally stable but evolutionary not-yet-observed protein folds, also called the 'dark matter' of protein structure space [68]. There is ....Pr ospects & Overviews A. Abroi and J. Gough a discrete-continuous duality of protein structure space (recently reviewed by Grishin and coworkers [69]), which is based on knowledge that the space is largely discrete in the evolutionary sense, but continuous geometrically.…”