The exploding fields of biomedical research, biotechnology and, more recently, synthetic biology require the development of computational techniques paralleled by experimental methods to efficiently design novel materials inspired by living systems. Proteins immobilized on functionalised polymeric or biological surfaces are of particular importance as they are routinely being utilized in numerous biological measurements in laboratory and clinical settings.
A better understanding of the self-assembly of highly ordered peptide nanostructures is vital as not only does it help to uncover the pathogenesis of various neurodegenerative diseases  but it also provides new clues for a "bottom-up" design  and fabrication of nanoscale devices and sensors  for use in biomedicine. In addition to a series of factors previously discovered to regulate peptide self-assembly [4-6], solid substrates have been found to drive this process directly acting as templates [7,8], controlling both assembly kinetics and morphology of amyloid peptide aggregates [9-11]. These observations emphasize the importance of the nature of the substrate/interface, whether it be a cellular membrane or an inorganic solid surface or something in between, on the assembly of various peptides, which is critical in many biological and nanotechnological processes. For example:
The self-assembly of peptides on titanium interfaces have recently been shown to enhance biological conjugation of implants .
Immobilizing membrane proteins at interfaces for screening of pharmaceutical targets, characterizing their structure and function & development of biosensors .
Label-free protein detection via the self-assembly of protein-metallic nanoparticle structures .
Formation of patterned surfaces on solid substrates using self-assembled proteins .
One of the primary challenges that is faced when designing these nanoscale devices and sensors is the ability to tune the interface in order to optimise the self-assembly of the peptides near the device interface for the desired application. Therefore obtaining a molecular scale understanding of the interactions between peptides and the desired interface is of utmost importance. Computer simulations have proven to be a very powerful tool in providing insight into these interactions and therefore will continue to play a significant part in the further development and design of these systems [16,17].
Due to the broad range of applications driven by this science, this field continues to become more and more interdisciplinary including medics, biologists, chemists, physicists, material scientists and engineers. Therefore, we propose to organise a CECAM workshop on this very exciting topic in which we will bring together the pre-eminent scientists in the various fields to discuss the open questions from the experimental and computational angles of attempting to gain a new level of understanding of how functionalised interfaces interact with the aggregation of proteins.
1) Parametrization of interface/protein interactions. While there is a lot of interest in understanding the molecular interactions which govern protein aggregation at functionalized interfaces, this is also an area with some significant challenges for the computer simulation community to overcome in order to be able accurately answer these questions. Maybe the most pressing issue is how to accurately model the interactions between the substrate and the peptide containing solution. Often times simulations of protein adsorption onto functionalized interfaces are carried out with parameters from the same classical force field (i.e. CHARMM) assuming that they will accurately reproduce the binding energy of the peptide onto that surface. However, it has been shown that in fact the binding energies of individual amino acids can be significantly far away from experimentally observed values in such a scenario [18,19]. Therefore more thoughtful approaches need to be adopted in order to model these systems such as the Interfacial Force Field  for modelling the interactions between the solution and substrate or creating general transferable force field parameters that accurately model the interaction between a common substrate and proteins (as has been done for example with GolP for gold substrates ).
2) Sampling of Initial & Final Configurations. Once the interactions are modeled correctly, then the method used to actually carry out the simulations and predicting the actual minimum energy structure of the protein at the interface is another challenge. While currently a lot of scientists carry out multiple simulations in which they start with the protein of interest being rotated to different angles in relation to the substrate and then measure the minimum final system energy in order to determine the minimum energy structure. However, as the protein increases in size, the number of initial structures that are required grows significantly and therefore results in a method which may not be as efficient in finding the actual minimum energy structure. Therefore, a lot of research is being taken to investigate novel ways to sample configurations of adsorbed proteins including modified replica exchange , docking algorithms , and metadynamics .
3) Multiscale Force-Fields. Finally, a coarse-grain force field that can accurately model the adsorption of proteins to interfaces of different types as of yet does not exist. The most well known coarse grain protein forcefield is MARTINI and it does not allow secondary structure of the protein to change during the course of the simulation, and therefore the final structure of the adsorbed protein would be incorrect. However, this is an area of research where a variety of methods are being applied to overcome this limitation of the coarse-grain models [25,26], which will be required to carry this field of computational research from studying the smaller scale academic problems to the larger scale practical type of problems. Even more in its infancy is the development of force fields that are effectively multiscale, such that the different scales can be integrated in one simulation . Developments in this area would be instrumental for the processes studied here.