Natural proteins do not ever contain metal components as far as we know. Natural proteins are composed of natural amino acids which only contain H,C,O,N,S. Selenocysteine contains Se (also a non-metal!) but it's a protinogenic amino acid, which means its a precursor to proteins but doesn't typically show up in the protein itself. From the Wikipediate page that you gave us in your question: "Metalloprotein is a generic term for a protein that contains a metal ion cofactor" and "A cofactor is a non-protein chemical compound or metallic ion".

It is true that metals are typically harder to model than C,H,O,N,S, and even Se, if doing ab initio calculations on the metals or metail-containing complexes. However the purpose of machine learning in protein folding studies, is to skip ab initio, statistical-dynamical and/or molecular-dynamical calculations of the relevant structures and simply use training data to predict the protein structures. That being said, there needs to be enough training data available (as you correctly pointed out) to learn what happens near the metal co-factors: The answer to this is that there's indeed enough metalloproteins to sufficiently populate a training set, but they won't contain enough of the specific metals involved in every metalloprotein. For example lots of data will be available for proteins containing Fe since Fe is in hemoglobin (for example) which is essential to the functioning of red blood cells to absorb oxygen; but the protein vanabins contains vanadium which is much more rare and therefore training data involving it will be much less available. You're also correct that metal elements can form more bonds than typical elements found in organic compounds.

So it depends on the metal in the relevant co-factor. Fe-based co-factors will have quite a lot of training data available, as will Mg-based ones, Zn-based ones, and a lot of other ones which contain the "more common" metals. For proteins like vanabins which contains vanadium, you are quite correct that training data will be limited, but also keep in mind that vanabins is a very rare protein found in sea squirts and we already know more about its structure (through X-ray crystallography, which means we don't need machine learning for it) than we even know about what it even does. The chances of other vanadium-containing co-factors in metalloproteins being very significant is too low to justify working on protein folding algorithms specifically for them.

Natural proteins do not ever contain metal components as far as we know. Natural proteins are composed of natural amino acids which only contain H,C,O,N,S. Selenocysteine contains Se (also a non-metal!) but it's a proteinogenic amino acid, which means it's a precursor to proteins but doesn't typically show up in the protein itself. From the Wikipedia page that you gave us in your question: "Metalloprotein is a generic term for a protein that contains a metal ion cofactor" and "A cofactor is a non-protein chemical compound or metallic ion".

It is true that metals are typically harder to model than C,H,O,N,S, and even Se, if doing ab initio calculations on the metals or metail-containing complexes. However, the purpose of machine learning in protein folding studies is to skip ab initio, statistical-dynamical and/or molecular-dynamical calculations of the relevant structures and simply use training data to predict the protein structures. That being said, there needs to be enough training data available (as you correctly pointed out) to learn what happens near the metal co-factors: The answer to this is that there are indeed enough metalloproteins to sufficiently populate a training set, but they won't contain enough of the specific metals involved in every metalloprotein. For example, lots of data will be available for proteins containing Fe since Fe is in hemoglobin (for example) which is essential to the functioning of red blood cells to absorb oxygen, but the protein vanabins contains vanadium which is much rarer and therefore training data involving it will be much less available. You're also correct that metal elements can form more bonds than typical elements found in organic compounds.

So it depends on the metal in the relevant co-factor. Fe-based co-factors will have quite a lot of training data available, as well as Mg-based ones, Zn-based ones, and a lot of other ones which contain the "more common" metals. For proteins like vanabins which contains vanadium, you are quite correct that training data will be limited, but also keep in mind that vanabins is a very rare protein found in sea squirts and we already know more about its structure (through X-ray crystallography, which means we don't need machine learning for it) than we even know about what it even does. The chances of other vanadium-containing co-factors in metalloproteins being very significant is too low to justify working on protein folding algorithms specifically for them.