crystal structure widely available through a formal and accepted route clearly creates
some powerful opportunities to address a number of the problems we currently face.
Despite increasing digitisation and automation, in small molecule crystallography, it is unavoidable that an increase in the rate of data production requires input
from larger numbers of skilled people. The requirement for skilled intervention and
oversight in data management has been mentioned above, but there are other crucial
aspects that currently need more ‘pairs of hands’ to address the increase in data
collection rate. Many facilities need to optimise their operational time to be justifiable, and slightly perversely rapid data collections can be a problem here! Selection
of crystals remains a key, rate-determining stage, and while more effective and
automated screening of samples can now occur on the diffractometer, still finding
and mounting the most appropriate crystal can be laborious and technically quite
difficult, requiring numerous iterations and a lot of expert decision-making. Macromolecular crystallography tends to take the approach of measuring everything
because that can be done rapidly and then only using the best data. The diverse
nature of chemical samples makes this approach much more of a challenge to
automate; however, there are still aspects of the philosophy of it that could be
adopted in chemical crystallography.
An obvious way to optimise the efficiency of the modern crystallographic facility
is to have many more skilled users – it might be argued that the era of the ‘lone
service crystallographer’ is rapidly disappearing. In some cases, a large research
group will have its own instrument and everybody will use it – in much the way
macromolecular crystallography has run over the years. However, for the service
crystallography facility, this is not the case, but there is an increasing trend for users
of the service being sufficiently trained in order that they can perform the data
collection themselves. Here the role of the expert crystallographer is changing to be
that of an educator as well. However, while it is relatively easy to train a novice how
to operate a diffractometer, this by no means helps with the task of processing and
analysing the arising data. There are such a wide variety of difficult problems
(sometimes generated by collecting poor data) in structure refinement, many of
which are very difficult to generalise, that mean for the foreseeable future this aspect
of the process will be where the main bottleneck arising from increased rate of data
collection lies.
2.3.2 Quality of Data Generated
In the modern facility, the quality of data generated has a much greater spread than
historically. In past eras when instrument time was rare and precious and had its
limitations, there was a strong tendency to collect data on the better-/best-quality
samples only. Modern instrumentation removes much of the instrument time availability problem and in turn allows one to address more challenging problems.
Furthermore, there is a strong demand in certain currently fashionable areas of
chemistry for any kind of result to add to the body of evidence to support claims.
Also, the way in which the technique supports different disciplines has an effect on
100
S. J. Coles et al.
some powerful opportunities to address a number of the problems we currently face.
Despite increasing digitisation and automation, in small molecule crystallography, it is unavoidable that an increase in the rate of data production requires input
from larger numbers of skilled people. The requirement for skilled intervention and
oversight in data management has been mentioned above, but there are other crucial
aspects that currently need more ‘pairs of hands’ to address the increase in data
collection rate. Many facilities need to optimise their operational time to be justifiable, and slightly perversely rapid data collections can be a problem here! Selection
of crystals remains a key, rate-determining stage, and while more effective and
automated screening of samples can now occur on the diffractometer, still finding
and mounting the most appropriate crystal can be laborious and technically quite
difficult, requiring numerous iterations and a lot of expert decision-making. Macromolecular crystallography tends to take the approach of measuring everything
because that can be done rapidly and then only using the best data. The diverse
nature of chemical samples makes this approach much more of a challenge to
automate; however, there are still aspects of the philosophy of it that could be
adopted in chemical crystallography.
An obvious way to optimise the efficiency of the modern crystallographic facility
is to have many more skilled users – it might be argued that the era of the ‘lone
service crystallographer’ is rapidly disappearing. In some cases, a large research
group will have its own instrument and everybody will use it – in much the way
macromolecular crystallography has run over the years. However, for the service
crystallography facility, this is not the case, but there is an increasing trend for users
of the service being sufficiently trained in order that they can perform the data
collection themselves. Here the role of the expert crystallographer is changing to be
that of an educator as well. However, while it is relatively easy to train a novice how
to operate a diffractometer, this by no means helps with the task of processing and
analysing the arising data. There are such a wide variety of difficult problems
(sometimes generated by collecting poor data) in structure refinement, many of
which are very difficult to generalise, that mean for the foreseeable future this aspect
of the process will be where the main bottleneck arising from increased rate of data
collection lies.
2.3.2 Quality of Data Generated
In the modern facility, the quality of data generated has a much greater spread than
historically. In past eras when instrument time was rare and precious and had its
limitations, there was a strong tendency to collect data on the better-/best-quality
samples only. Modern instrumentation removes much of the instrument time availability problem and in turn allows one to address more challenging problems.
Furthermore, there is a strong demand in certain currently fashionable areas of
chemistry for any kind of result to add to the body of evidence to support claims.
Also, the way in which the technique supports different disciplines has an effect on
100
S. J. Coles et al.