The information necessary to generate many of the complex assemblies of macromolecules in cells must be incorporated in the subunits themselves. This is due to the fact that purified subunits are capable of spontaneously assembling into the final structure under the correct conditions. The first large macromolecular aggregation that was capable of self-assembling from its component parts was the tobacco mosaic virus (TMV). This virus takes the form of a long rod and has an RNA core that is encased in a protein cylinder at its center. When the RNA and protein pieces of the virus are brought together in a solution, they are able to recombine and form fully functional viral particles. During the course of the assembly process, surprisingly complex protein double rings are generated. These rings act as intermediates that contribute to the thickening of the viral coat.

Another example of a complex macromolecular aggregate that is capable of dismantling and reassembling itself is the ribosome, which is found in bacteria. This structure is made up of around 55 different types of protein molecules and 3 different types of rRNA molecules. When a mixture of the components is allowed to be incubated in a test tube under the appropriate conditions, the initial structure will automatically reform on its own. It is essential for restored ribosomes to retain their ability to catalyze the production of proteins. As one might assume, the reassembly of ribosomes occurs in a programmed manner: once specified proteins have joined to the RNA, this complex is then recognized by more proteins, and so on and so forth, until the structure is completed.

It is still not completely understood how some of the more complicated self-assembly processes are regulated. For example, many structures inside the cell appear to have precisely specified lengths that are noticeably longer than the lengths of the macromolecules that make up those structures. These lengths appear to be significantly greater than the lengths of the structures themselves. It is not always clear how such measurements of length are obtained. In the simplest case, a lengthy core protein or some other macromolecule serves as a scaffold and determines the scope of the final assembly. This process, which determines the length of the TMV particle, has its focal point at the RNA chain, which serves as the center of the process. It is believed that a core protein that interacts with actin is responsible for controlling the length of the thin filaments that can be found in muscle.

When it comes to holding self-assembling cellular structures together, noncovalent connections aren't always the best option. For example, a cilium or a myofibril of a muscle cell cannot form on its own out of a solution of the macromolecules that make them up. This is because the macromolecules are too large. In these circumstances, specialized enzymes and other proteins that function as templates and operate as assembly factors that direct construction but do not participate in the final formed structure offer some of the assembly information. These enzymes and proteins do not take part in the formation of the final structure. Even relatively simple constructions may be missing some of the components necessary for their own construction. This might cause serious problems. For instance, the head of certain bacterial viruses is created on a temporary scaffold made of a second protein that the virus creates. This head is made up of several copies of a single protein subunit. Because the final viral particle is missing the second protein, the viral head structure is unable to spontaneously reconstruct after being disassembled. This prevents the virus from spreading to other hosts.

One well-known instance of a step in the normal assembly process that is both required and irreversible is called proteolytic cleavage. Even some very modest protein assemblages, like collagen and insulin, display similar behavior. Insulin is a hormone, whereas collagen is a structural protein. It appears obvious from these relatively simple examples that the formation of a structure as complex as a cilium will undoubtedly involve a temporal and spatial ordering that is provided by a large number of other components. This can be deduced from the fact that cilia are found in all multicellular organisms.

When not under control, a particular class of protein structures that is used for several typical cellular processes can also be a factor in human disorders. These disorders can be caused by the fact that this class is not under control. These sheet aggregates, which are also referred to as amyloid fibrils, are stable and capable of self-replication. These fibrils are constructed from a continuous stack of sheets made of identical polypeptide chains that are stacked one on top of the other to produce a cross-beta filament. The sheets are stacked in such a way that the strands are oriented perpendicular to the axis of the fibril. In most cases, the combination of hundreds of monomers will result in the formation of a fibrous structure that is unbranched and has dimensions ranging from 5 to 15 nm in width and many micrometers in length. Due to the fact that the short polypeptide chain that makes up the spine of the fibril can have a range of different sequences and follow one of several different routes, a surprisingly significant percentage of proteins have the capacity to create structures of this kind.

In healthy individuals, the quality control mechanisms that govern proteins gradually weaken with increasing age, which occasionally makes it possible for normal proteins to gather in a pathogenic manner. After the protein aggregates have been released by the dead cells, the extracellular matrix may eventually form amyloid from them. In extreme cases, the accumulation of such amyloid fibrils within the cells can result in harm to the tissues as well as the death of the cells. Because the brain is made up of a highly organized collection of nerve cells that do not have the ability to regenerate, it is especially vulnerable to this form of harm that accumulates over time. Therefore, although amyloid fibrils can form in a variety of tissues and are known to produce pathologies in a number of places throughout the body, the most severe amyloid pathologies are neurodegenerative illnesses. For instance, the abnormal formation of very stable amyloid fibrils is thought to play a significant role in the pathogenesis of both Alzheimer's disease and Parkinson's disease.

Prion disorders are a particular subset of this larger group of ailments. Prion diseases, in contrast to Parkinson's disease and Alzheimer's disease, can be passed on from one creature to another if the second organism consumes a tissue that carries the protein aggregation. This is one of the reasons why prion diseases have earned such a prominent position in recent years. Creutzfeldt-Jakob disease (CJD) in people, also known as Kuru, and bovine spongiform encephalopathy (BSE) in cattle are all examples of a group of diseases that are closely related to one another and are brought on by an abnormally folded and aggregated form of a particular protein known as PrP. In most cases, PrP may be located on the exterior of the plasma membrane, and this is especially true in neurons. Unfortunately, PrP possesses a terrible ability to form amyloid fibrils, which are regarded as "infectious" due to the fact that they transform normally folded molecules of PrP into the same diseased form. Amyloid fibrils are associated with Alzheimer's disease. Because of this property, the incorrect conformation swiftly spreads from one cell to the next in the brain, which finally results in death and propagates the abnormal form of PrP known as PrP*. This condition ultimately leads to death. Consuming the tissues of animals that are infected with PrP can put one at risk of contracting the disease, as demonstrated by the transmission of BSE, sometimes known as "mad cow disease," from cattle to people. When PrP* is missing, it is very difficult to convert PrP into its abnormal form. This is a fortunate circumstance.

A phenomena called "protein-only inheritance" has been observed to occur in yeast cells, which is directly related to the topic. Because infectious proteins can now be studied in yeast, yet another puzzling element of prions has been brought to light. These protein molecules are capable of generating a wide variety of amyloid fibrils, each of which is distinct from the others, from a single polypeptide chain. Additionally, there is the possibility that each type of aggregate will expand, leading to ordinary protein molecules adopting the same peculiar shape. Therefore, several "strains" of infectious particles are capable of developing from a single polypeptide chain.