How can these approaches be integrated with experiment

The outlined methods for assembly of protein models have important applications to structure determination from sparse experimental data. The most is to model building from rather complete NMR data. Work by Bayley and coworkers [193] is a good example of a novel approach to this well-defined and standard problem. 

For some cases, it is easier to experimentally determine the secondary structure of a native protein without any information about tertiary interactions. The work described in Section 2.2 is aimed at structure determination from just knowledge of secondary structure [48, 49, 159, 164, 177, 188-190]. For regular helical proteins, these methods may provide low-resolution models. For more complex topologies, the probability of building a good molecular model just from secondary structure information is rather small. 

---