The international bibliography of electron microscopy.

2

Raman microscopy of the molybdate minerals koechlinite, iriginite and lindgrenite

A series of molybdate bearing minerals including wulfenite, powellite, lindgrenite and iriginite have been analysed by Raman microscopy. These minerals are closely related and often have related paragenesis. Raman microscopy enables the selection of individual crystals of these minerals for spectroscopic analysis even though several of the minerals can be found in the same matrix because of the paragenetic relationships between the minerals. The molybdenum bearing minerals lindgrenite, iriginite and koechlinite were studied by scanning electron microscopy and compositionally analysed by EDX methods using an electron probe before Raman spectroscopic analyses. The Raman spectra are assigned according to factor group analysis and related to the structure of the minerals. These minerals have characteristically different Raman spectra.

Automatic particle picking of biological molecules imaged by electron microscopy

One of the next great challenges of cell biology is the determination of the enormous number of protein structures encoded in genomes. In recent years, advances in electron cryo-microscopy and high-resolution single particle analysis have developed to the point where they now provide a methodology for high resolution structure determination. Using this approach, images of randomly oriented single particles are aligned computationally to reconstruct 3-D structures of proteins and even whole viruses. One of the limiting factors in obtaining high-resolution reconstructions is obtaining a large enough representative dataset ($>100,000$ particles). Traditionally particles have been manually picked which is an extremely labour intensive process. The problem is made especially difficult by the low signal-to-noise ratio of the images. This paper describes the development of automatic particle picking software, which has been tested with both negatively stained and cryo-electron micrographs. This algorithm has been shown to be capable of selecting most of the particles, with few false positives. Further work will involve extending the software to detect differently shaped and oriented particles.