Please see the Newsletter Article for complete details. The purpose of this page is to give a quick summary sheet of the command line options in the current release, showing how they relate to the original SSRL distribution, as well as the internal source code variables.
usage: distl.signal_strength image_filename [parameter=value ...] example: distl.signal_strength lysozyme_001.img distl.res.outer=2.0 distl.res.inner=6.0 distl.minimum_spot_area=8
| Command line option & Default | Description | Legacy flag | libdistl variable | |
| distl.image=None | Image file name | |||
| distl.res.outer=None | High resolution limit in angstroms | |||
| distl.res.inner=None | Low resolution limit in angstroms | |||
| distl.verbose=False | Lengthy spot printout | |||
| distl.minimum_spot_area=None | Override application default (PADs:5, others:10) and set minimum spot area (in pixels) within spotfinder. | -s2, -s3 | The two corresponding variables (spotarealowcut and spotbasesize) are both set to the value | |
| distl.minimum_signal_height=None | Override application default (PADs:2.5, CCDs:1.5) and set minimum signal height (in units of background noise sigma) within spotfinder. | -bg0, -bg1, -bg2 | bgupperint[0],bgupperint[1],bgupperint[2] are all set to the value. | |
| distl.minimum_spot_height=None | Expert use only; after pixels are classified as signals rather than noise, minimum height to be considered a spot maximum (in units of background noise sigma). Default=3.5 | -d1 | difflowerint | |
| distl.spot_area_maximum_factor=None | Expert use only; max spot area expressed as a multiple of minimum_spot_area. Default=5.0 | -s7 | spotareamaxfactor | |
| distl.peak_intensity_maximum_factor=None | Expert use only; peak intensity filter used by libdistl. Default=10.0 | -s8 | spotpeakintmaxfactor | |
| distl.scanbox_windows=101,51,51 | Expert use only; Integer scanbox sizes (for calculating background) for cycles 1, 2, and 3, respectively. The program defaults are 101, 51, and 51 pixels; square boxes have this number of pixels on an edge. | -bx0, -bx1, -bx2 | scanboxsize | |
| distl.peripheral_margin=20 | No spot detection within a peripheral margin of this width (in pixels) | imgmargin | ||
| distl.pdf_output=None | File name for optional PDF graphical output for distl.signal_strength (*.pdf). Requires PHENIX. pink=max_pixel; red=center_of_mass | |||
| distl.image_viewer=False | Start up the interactive image viewer to see the spotfinder chosen spots. | |||
| distl.bins.verbose=False | Additional printout binned by resolution range. | |||
| distl.bins.N=20 | Maximum number of bins, but fewer can result if there are few spots. | |||
| distl.bins.corner=True | Extend the binning all the way to detector corner, otherwise to outermost spot on first image. | |||
| distl.port=8125 | For the server version, port number to listen for requests | |||
| distl.processors=1 | For the multithreaded server version, number of server processes to be used |
Verbose analysis of spots, broken down by resolution bin, can be obtained by adding the following command line options:
distl.signal_strength distl.bins.verbose=True distl.bins.N=[number of bins] /home/data/user/PI09_1_E1_001.imgThe following information is tabulated in the verbose output:
| Limit | The outer edge of the (equal-volume) resolution shell (Angstroms). |
| Population | The number of spots in each resolution shell. |
| Missing | The fraction of that resolution shell that would be recorded by a 360-degree rotation, accounting for the missing cone. |
| Fract | The fraction of that resolution shell recorded on an image, accounting for resolution shells that are geometrically outside the detector edge, but are recorded in the detector corners. |
| PxlBkgrd | The mean pixel background at center of the scanbox windows in each resolution shell. |
| MnWndwSz | The mean scanbox window size in square pixels after discarding pixels classified as signals (such as Bragg spots). |
| Integrated | The total integrated signal (for all spots) within the shell in analog-digital units above background. |
| MeanI | The mean integrated signal for spots within the shell. |
| MeanBkg | The mean integrated background for spots within the shell. |
| MeanSz | The mean number of pixels for spots within the shell. |
| MeanIsigI | The mean[I / sig(I)] for spots within the shell. Sigma is defined as sqrt(variance) with variance being the variance in I + variance in Bkg + variance in the WndwSz pixels used to compute background. Variances add in quadrature as in Acta Cryst D55, p 1696 eq. 9 (Andrew Leslie) and the gain is taken as 1.0. |
| MnEccen | The mean eccentricity for spots within the shell; 0=circle, 1=parabola. |
| MnSkew | The mean skewness for spots within the shell; skew:= (maximum-center_of_gravity)/semimajor_axis. |
Example text output is:
Analysis of spots after ice removal, but prior to resolution cutoff, for image PI09_1_E1_001.img:
Limit Population Missing Fract PxlBkgrd MnWndwSz Integrated MeanI MeanBkg MeanSz MeanIsigI MnEccen MnSkew
2.33 871 0.99 1.00 104.0 597 5986386 6873 7991.0 67 37.906 0.825 0.206
1.85 53 0.97 1.00 67.9 582 42144 795 2618.4 37 11.831 0.842 0.230
1.61 44 0.95 0.40 56.4 696 5752 131 728.7 13 4.267 0.847 0.258
1.47 7 0.94 0.14 53.9 618 436 62 648.2 12 2.304 0.813 0.227
---------- ----------
975 6034717 6189 7318.5 63 34.715 0.826 0.210
Analysis of spots after resolution filtering, but prior to spot quality heuristics, for image PI09_1_E1_001.img:
Limit Population Missing Fract PxlBkgrd MnWndwSz Integrated MeanI MeanBkg MeanSz MeanIsigI MnEccen MnSkew
2.33 648 0.99 1.00 104.0 597 5495315 8480 8802.3 71 44.186 0.814 0.202
---------- ----------
648 5495315 8480 8802.3 71 44.186 0.814 0.202
Analysis of good Bragg spots after quality heuristics, for image PI09_1_E1_001.img:
Limit Population Missing Fract PxlBkgrd MnWndwSz Integrated MeanI MeanBkg MeanSz MeanIsigI MnEccen MnSkew
2.33 312 0.99 1.00 104.0 597 2529932 8109 6271.6 50 45.496 0.753 0.137
---------- ----------
312 2529932 8109 6271.6 50 45.496 0.753 0.137
File : PI09_1_E1_001.img
Spot Total : 986
Remove Ice : 975
In-Resolution Total : 648
Good Bragg Candidates : 312
Ice Rings : 0
Method 1 Resolution : 2.24
Method 2 Resolution : 3.41
Maximum unit cell : 272.1
<Spot model eccentricity> : 0.762
%Saturation, Top 50 Peaks : 4.32
In-Resolution Ovrld Spots : 0
Bin population cutoff for method 2 resolution: 20%
Number of focus spots on image #1 within the input resolution range: 648
Total integrated signal, pixel-ADC units above local background (just the good Bragg candidates) 2529931
Signals range from 234.2 to 228761.6 with mean integrated signal 8480.4
Saturations range from 0.0% to 13.6% with mean saturation 0.5%
As of 19 July 2011, the following additional options can be included on the distl.signal_strength command line:
| Command line option & Default | Description | Expert level |
| autoindex_override_beam=None | x and y coordinates of the direct beam in mm (two values separated by comma). | Basic |
| autoindex_override_distance=None | Crystal-to-detector distance in mm. | Basic |
| autoindex_override_wavelength=None | Incident wavelength in Angstroms. | Basic |
| autoindex_override_twotheta=None | Detector swing angle in degrees. | Basic |
| autoindex_override_deltaphi=None | Single-shot rotation angle in degrees. | Basic |
| image_specific_osc_start=None | A Python lambda x expression giving the rotation in degrees, given the image number; for example "lambda x: x-1.0" signifies 0 degrees for image 1; 1 degree for image 2, etc. | Basic |
| wedgelimit=2 | Maximum number of images to use for Labelit indexing [does not affect Spotfinder directly]. | Advanced |
| spotfinder_verbose=True | Set Spotfinder to print software development messages. | Advanced |
| tnear_resolution_divisor=5 | Control of the cutoff for determination of resolution by Method 2. A lower value (1.1) makes a more stringent cutoff, so the outer resolution cutoff is at a higher d-spacing (lower resolution). A higher value (10) is more permissive and allows higher resolution Bragg spots, but can cause Labelit indexing failure in some cases. | Advanced |
| force_method2_resolution_limit=None | Override the normal Method 2 resolution analysis based on the falloff of spot count. Force spots to be chosen to this limit (Angstroms). | Advanced |
| distl_binned_image_spot_size=4 | Special case; hardcoded algorithm. For CCD-based area detectors with large pixel count (>4000-squared); images are binned 2x2 in software prior to spot analysis, and minimum spot area set to this number of binned pixels. | Advanced |
| distl_profile_bumpiness=2 | Maximum number of local maxima permitted in good Bragg spots. It is necessary to set this parameter higher for home-lab sources with double-focusing mirrors. Contact authors with questions. | Advanced |
| distl_permit_binning=True | Permit 2x2 binning for large images (>4000 pixels-squared). | Advanced |
| distl_force_binning=False | Force 2x2 binning for all images; only used for development and troubleshooting. | Advanced |
| codecamp.maxcell=None | Directly specify the maximum unit cell (based on previous knowledge of the molecule); new algorithm to permit Labelit indexing on contiguous images. | Advanced |
| codecamp.minimum_spot_count=None | For determining spot masks on single images, minimum allowable spot count. | Advanced |