Access for data collection
Any research group, institutes and companies, who would like to use the National Electron CryoMicroscopy (NECM) facility to collect EM images should submit a proposal, which will undergo peer-review process. After the proposal has been evalauted by reviewers, time will be allocated typically in the beginning of the week. The data will be collected by a NECM microscope operator, in the presence of a person from the particular research group or company. Only those who have had formal traning will be allowed to mount the grids and operate the microscope. The collected data will be transferred to a portable hard drive allowing users to process the data at leisure at home. We will be happy to provide advise on data processing but currently we will not be able to give access for data processing.
As explained in the intro page, a wide variety of specimen can be imaged with EM at cryo temperatures. It is important to remember that the quality of the specimen (in case of single particle cryoEM, biochemistry of the protein) and the grid preparation largely determines the outcome of the data collection. A biochemically well characterised protein solution will in most cases result in very good grids and images. Thus, the data collection and processing becomes easier. Heterogeneity per se is not a big issue as classification during processing can be used to sort through different populations but if this can be addressed biochemically then it saves time. Our advise is to be realistic and very often one finds that a sample that gives a good gel filtration profile or a single band in gel might look completely different on a micrograph. Please do not blindly follow the protocols described in publications but try to explore conditions with the protein of your choice.
1) The amount of protein required for making an EM grid will depend on many factors (including the size of the protein, if a carbon or graphene layer is used). For a rough estimation of how much protein is required to image in ice, please refer to the table in the review - Vinothkumar & Henderson, Quarterly Review of Biophyscis, 2016.
2) High-resolution structures of small proteins such as Haemoglobin (64 kDa) has been determined by single particle EM and this lower limit is likely to reduce as we understand more about imaging and technology develops . While such moleuclar sizes is doable, such projects take a lot of effort and when possible crystallization and X-ray Crystallography should be carried out in parallel.
3) More images or particles doesn't guarantee increase in resolution. Often one would find that the resolution is limited due to the nature of protein in particular as many are dynamic. As an initial step it is useful to collect a small to medium size data set (no more than 700 images or 30-40,000 particles), check if the refinement and reconstruction goes well. Using the B-factor estimated from the data (see Rosenthal and Henderson 2003), it is possible to estimate how many more particles you will need to attain the resolution of your interest. We highly recommend this before collecting large data sets, which will require a number of days of computing yet may not result in higher resolution.