INTRODUCTION Crystallography was in the beginning a domain of mineralogy and mathematics

INTRODUCTION
Crystallography was in the beginning a domain of mineralogy and mathematics, mostly dealing symmetry and crystal lattices. Wilhelm Conrad Röntgen discovered X-rays in 1895 and later on, in 1912 it was discovered that X-ray irradiated salt crystals produce diffraction patterns, thus revealing the internal atomic periodicity of the crystals. About 20 years later, protein crystallography started, uncovering a new era of achievements in the next decades, by developments in both the theoretical and technical sides.(1)
A BRIEF HISTORY
X-ray sources
Many years sealed tubes were the standard laboratory X-ray sources. A lamp filament shaped cathode emitted electrons that were accelerated to energies of hundreds of kilovolts, and the rays were produced by their impact on metal anodes. Later, rotating-anodes started to be used, as they could resist more and give more intense rays. The last ones still remain used nowadays where synchrotron is not accessible. Synchrotron is one of the newer technologies, as it was proven that it can be used as a much stronger X-ray source, thus generating a higher resolution.(1)
Early on, photographic films and precession cameras were the ways of collecting the data. Given the weak diffraction of protein crystals and their faster decay, it is unquestionably necessary to have a simultaneous collection of the weak diffraction spots spread along a large area together with a high spatial resolution. This was accomplished using X-ray-sensitive film. Nowadays, as a result of the cutting-edge developments in the domain, the most widespread methods for PX (protein crystallography) data-collection are CCDs (charge-coupled devices) and CMOS (complementary metal oxide semiconductor).(1)
THE FUTURE OF CRYSTALLOGRAPHY IN DRUG DISCOVERY
The 3D structures of biological macromolecules are considered the uppermost standard of characterizing their architecture. The obtained structures can give a richness of information for drug discovery attempts utilizing CADD (computer-aided drug design). (2)
Although there are many methods for determining molecular structures, X-ray crystallography is especially good one for drug discovery, because this method is capable of producing very high resolutions, even atomic-scale ones and this method enables the mapping of large heteromeric complexes, such as ribosomes. Furthermore, and one of the most important advantages is that X-ray crystallography can grant evidence of the binding way of small ligands in the crystal.(2)
The amount of macromolecular structures available is growing exponentially. The wholly use of the flourishing structure data, along with the abundance of other types of biological data available (such as amino acids, metabolic pathways, expression patterns) is a substantial challenge for data mining in a variety of medical applications, such as drug design.(2)
Although the structures deposited to the Protein Data Bank (PDB) are generally high quality ones, not all structural details are optimum. The precise spatial coordinates of all the atoms in a structure can be derived exclusively from the experimental maps for high-resolution structures. If the structure has a lower resolution, previous information, like chemical identity and stereochemistry, must be used to figure out the atoms positions in the molecule.(2)
CRYSTALS STABILISATION
While crystallizing it is crucial that the species remain the same within the crystallization experiment, thus some form of compositional stability is needed. The homogeneity of a specimen is generally determined by electrophoresis or mass spectroscopy. This does not guarantee that if, for example, the protein is stable as to lead to a single band on a SDS-PAGE electrophoresis, it will be sufficiently stable over the period of a crystallization experiment.(3)
Still, even if the protein sample has a considerable amount of compositional homogeneity, it is expected that unless it shows conformational stability, it will not crystallize, even if the compositional one is perfect. Many proteins show little conformational order. A big category of proteins, called intrinsically disordered proteins (IDPs) have considerable levels of conformational disorder, and sometimes even no appreciable 3D structure. These proteins are generally avoided by crystallographers, these structures being unveiled by NMR spectroscopy.(3)
After a crystal is formed, it needs to be harvested. Harvesting a crystal means to move the crystal from its growth place to the X-ray generating apparatus, which will collect the diffraction patterns. Generally this process is composed on multiple parts, like looping and cryoprotection, and finally putting it on the X-ray source.(4)
ADVANCES IN THE FIELD
Selenium was shown to have a use as an anomalous scatterer. This has been a huge progress as it made multi-wavelength anomalous dispersion (MAD) the main technique for determining new protein structures. The advantage is that the sulfur in methionine can be replaced with selenium generating selenomethionine, thus the element being incorporated in the protein. For nucleic acids, bromine can be used to modify the nitrogenous bases, having a similar effect as selenium for proteins.(5)
TIME RESOLVED X-RAY
Because biological macromolecules are dynamic systems, knowing the static three-dimensional structures is not always enough. Time-resolved crystallography produces some sort of films of the proteins’ behavior. One of the methods is the Pump-Probe Method, where a short laser pulse initiates a reaction. After a previously known time, an X-ray beam is sent, enabling us to see the reaction status at that given moment. Then, the sequence is repeated. In other situations, the laser pulse is followed by subsequent X-ray beams, enabling us to see the protein’s state at more than one given time.(6)
CRYSTALLIZATION
Crystallization of biomolecules is done in 2 indivisible steps: nucleation and growth. Nucleation, being a transition from a disordered state to a ordered state of a molecule, is a very difficult problem to solve. The growth mechanisms are much better known, and protein crystals are made primarily by dislocation growth and two-dimensional nucleation growth.(7)