data becomes unwieldy very rapidly – it does not scale well. Secondly, the nature of
the small molecule result is very different to that of a macromolecular one – the
resolution of protein structures is relatively low and there is often ‘room for
improvement’, whereas the likelihood that relatively little extra can be obtained
from a high-resolution small molecule structure is much higher. As methodology
develops, in some cases very rapidly, the ability to reassess old data and get a better
result becomes very real and very worthwhile. There are numerous situations where
recourse to raw data could therefore be of very real benefit to future users of our data,
many of which are listed below:
• When a result provides a contribution to chemical knowledge, but is poor quality
• In order to support a ‘grand’ claim, i.e. an unusual result that has not been
previously observed or considered possible
• To support cases where modelling of disorder, twinning, incommensurate or
modulated structures could be open to other interpretations
• To support analysis of the modelling of diffuse scattering
• To make available, e.g. disorder, twinning, incommensurate, modulated, diffuse
scattering datasets so others/future generations can attempt to (re)solve them
• To provide supporting evidence for ‘Advanced Experiments’, e.g. charge density,
high pressure, phase transition, gas environment and excited states
• When it is clear that future improvement, e.g. for many of the cases mentioned
above, may be possible through developments in software and modelling
• To make available training sets and benchmarks for software and method
developers
As access to fast networks and large volumes of data storage become ubiquitous
and normal, many of the problems that have historically been perceived as blocks to
routinely curating, making available and accessing raw data have gone. As mentioned previously that is not necessarily a case for making everything available but
certainly is when to do so is clearly going to be of value.
Coupled with the recent technological developments described in Sect. 1, these
digital developments have provided a platform that is fundamentally changing the
application of structural information. Partly due to the fact that the technique is seen
as a much more rapid and accessible method of characterisation by colleagues in
synthesis and partly because specialists can achieve so much more, the number of
articles containing crystallographic information has risen dramatically. The Crystallographic Information Framework is very much an enabler for this phenomenon, and
the result is that being able to process and communicate at these new levels has led to
entirely new and increasingly independent fields of study.
2.2.1 Structure-Driven Independent Research Fields
The speed, accessibility and capability of the technique have led to many more
academic research groups being driven by crystallography and becoming almost
wholly dependent on it. This has given rise to fields such as crystal engineering and
96
S. J. Coles et al.
the small molecule result is very different to that of a macromolecular one – the
resolution of protein structures is relatively low and there is often ‘room for
improvement’, whereas the likelihood that relatively little extra can be obtained
from a high-resolution small molecule structure is much higher. As methodology
develops, in some cases very rapidly, the ability to reassess old data and get a better
result becomes very real and very worthwhile. There are numerous situations where
recourse to raw data could therefore be of very real benefit to future users of our data,
many of which are listed below:
• When a result provides a contribution to chemical knowledge, but is poor quality
• In order to support a ‘grand’ claim, i.e. an unusual result that has not been
previously observed or considered possible
• To support cases where modelling of disorder, twinning, incommensurate or
modulated structures could be open to other interpretations
• To support analysis of the modelling of diffuse scattering
• To make available, e.g. disorder, twinning, incommensurate, modulated, diffuse
scattering datasets so others/future generations can attempt to (re)solve them
• To provide supporting evidence for ‘Advanced Experiments’, e.g. charge density,
high pressure, phase transition, gas environment and excited states
• When it is clear that future improvement, e.g. for many of the cases mentioned
above, may be possible through developments in software and modelling
• To make available training sets and benchmarks for software and method
developers
As access to fast networks and large volumes of data storage become ubiquitous
and normal, many of the problems that have historically been perceived as blocks to
routinely curating, making available and accessing raw data have gone. As mentioned previously that is not necessarily a case for making everything available but
certainly is when to do so is clearly going to be of value.
Coupled with the recent technological developments described in Sect. 1, these
digital developments have provided a platform that is fundamentally changing the
application of structural information. Partly due to the fact that the technique is seen
as a much more rapid and accessible method of characterisation by colleagues in
synthesis and partly because specialists can achieve so much more, the number of
articles containing crystallographic information has risen dramatically. The Crystallographic Information Framework is very much an enabler for this phenomenon, and
the result is that being able to process and communicate at these new levels has led to
entirely new and increasingly independent fields of study.
2.2.1 Structure-Driven Independent Research Fields
The speed, accessibility and capability of the technique have led to many more
academic research groups being driven by crystallography and becoming almost
wholly dependent on it. This has given rise to fields such as crystal engineering and
96
S. J. Coles et al.