02034nas a2200361 4500008004100000022001400041245010100055210006900156260001200225300001100237490000800248520093900256653001501195653002001210653002801230653005301258653004001311653003101351653001301382653001801395653001401413653002601427653001301453653002401466100001701490700001701507700002101524700001801545700001901563700001901582700002301601856004801624 2005 eng d a1047-847700aParticle picking by segmentation: a comparative study with SPIDER-based manual particle picking.0 aParticle picking by segmentation a comparative study with SPIDER c12/2005 a211-200 v1523 a
Boxing hundreds of thousands of particles in low-dose electron micrographs is one of the major bottle-necks in advancing toward achieving atomic resolution reconstructions of biological macromolecules. We have shown that a combination of pre-processing operations and segmentation can be used as an effective, automatic tool for identifying and boxing single-particle images. This paper provides a brief description of how this method has been applied to a large data set of micrographs of ice-embedded ribosomes, including a comparative analysis of the efficiency of the method. Some results on processing micrographs of tripeptidyl peptidase II particles are also shown. In both cases, we have achieved our goal of selecting at least 80% of the particles that an expert would select with less than 10% false positives.
10aAlgorithms10aAminopeptidases10aCryoelectron Microscopy10aDipeptidyl-Peptidases and Tripeptidyl-Peptidases10aImage Processing, Computer-Assisted10aImaging, Three-Dimensional10aInternet10aParticle Size10aRibosomes10aSerine Endopeptidases10aSoftware10aSoftware Validation1 aAdiga, Umesh1 aBaxter, Bill1 aHall, Richard, J1 aRockel, Beate1 aRath, Bimal, K1 aFrank, Joachim1 aGlaeser, Robert, M uhttp://www.ncbi.nlm.nih.gov/pubmed/16330229