1"Bioc585 2008 Exercise 2 – Integrating Intensities Homework due Tuesday, April 15 We will use the programs iMOSFLM and the program suite CCP4i to integrate intensities from a cubic crystal of insulin, scale the intensities to minimize systematic errors, and convert the intensities to structure factor amplitudes. 1. Log in to your Linux account, open a terminal window, create a new directory for exercise two and start iMOSFLM. mkdir ex2 (creates a new directory) cd ex2 (changes your working directory to ex2) imosflm (initiates program – takes a few seconds) 2. Read in diffraction images. session menu > add images navigate to images folder: shared/585xray/images_insulin select first image and click open: all 45 images are read in 3. Explore image 1. use the +/- (zoom) buttons and the panning tool to examine the image. Note that the image file is simply a list of pixel values. click the “show resolution” button (sixth button from left). This shows the resolution limits at 1.876 Å. Select the images menu > View > Processing options > Processing. Change the resolution. What is the approximate resolution of these data? That is, how far out do the reflections go? Select an appropriate resolution and close the window.2" 4. Index the reflections. Click the indexing button to open the Autoindexing window, which defaults to using images 1 and 45. Click the index button. Multiple solutions to the indexing problem should appear, all with a Bravais Lattice (ignore for our purposes), an overall error (Pen.), cell constants, and errors in the matching of observed and calculated reflection positions (both in x,y position and in angular degrees). Examine an image: the blue boxes are predicted positions. The best choice for unit cell is highlighted (solution 12). In general, one chooses the solution that has the most symmetry while maintaining low error. Examine the possible space groups consistent with the solution 12: four are listed. Examine the four space groups to determine the number of molecules in the cell and any systematic absences. Is there any way to distinguish between the four based solely on the diffraction pattern? 5. Estimate the mosaicity. This is the reflection width in degrees. How wide are the reflections on average? 6. Examine Strategy. This defines which part of the possible 360° are needed to obtain a complete data set with good symmetry redundancy. Since the predicted cell is cubic, only a small wedge is needed. 7. Cell Refinement. Select the Cell Refinement window and click “Process”. The initial predicted cell and camera parameters must be adjusted to get the best fit between predicted and observed reflections. Note the final cell constants. How much did the unit cell change by? What is the final residual? Note orientation matrix that rotates cell axes onto camera axes (phi"angles).""8.#Integration.##Integrate"all"45"images.""Note"how"well"the"reflections"fit"in"the"integration"boxes" and" error" levels." " The" integrated" reflections" are" in" a" binary" file" called"insulin_1_001.mtz"if"you"chose"the"default"file"name.""Details"concerning"the"integration"are"under"history">"log,"but"only"for"each"image.""""3"9.# Exit# iMOSFLM,# start# CCP4i." " Exit" by" quitting" in" the" program" menu." " Type" “ccp4i”" to"begin" the" new" programs." " Choose" “Directories" &" Project" Directory”." " Change" the" project"name" to" insulin" and" change" the" directory" to" your" current" directy" (e.g." ex2)." " Select" the"“Project"for"this"session”"as"insulin"(bottom"of"page)"and"click"“apply"&"exit”.""10.# Cell# volume." " The" cell" volume" and" the" protein" volume" can" be" estimated" to" give" an"estimated" percentage" solvent" for" the" crystal," assuming" one" or" more" molecules" per"asymmetric"unit.""A"typical"protein"crystal"has"~50%"solvent,"but"can"vary"widely.""Select"“Coordinate" Utilities”" >" “Cell" Content" Analysis”." " Type" in" the" cell" parameters" and" the"possible"unit"cells.""Insulin"has"51"residues,"so"the"molecular"weight"is"~5500"Da.""Can"you"distinguish" whether" the" asymmetric" unit" has" 1" or" 2" molecules" in" this" way?" " Can" you"distinguish"between"space"groups?"""11.# Scale# Intensities# and# convert# to# Structure# Factor# Amplitudes." " Select" “Data"Reduction”" >" “Scale" and" Merge" Intensities”." " Select" your" .mtz" file," select" “Run" Truncate”,"select"“Ensure"unique"data"&"add"FreeR”"and"click"Run.""Close"the"window."##4"In" the" main" window," wait" until" the" job" has" finished" and" examine" the" log" graphs." " In"particular," examine" completeness," I/sigma(I)," Rmerge" and" redundancy." " Run" the" mouse"over"each"data"point"to"display"its"value."""""" """Homework Questions. 1) Estimate the resolution of the data from the images, using the “show resolution” button. 2) Is there any way to distinguish between the four possible space groups based solely on the diffraction pattern? 3) Is there any way to distinguish between the four possible space groups based on cell content? 4) What were the final Rmerge values at low and high resolutions? 5) What were the final I/sigma(I) values at low and high
View Full Document