SLIDE 1
Idealized bio-SAS experiment
Solution Scattering Data from Protein of Interest Black Box Neutron source/beamline homesource
Automated data analysis on ESRF BM29 Martha Brennich (EMBL Grenoble) - - PowerPoint PPT Presentation
Automated data analysis on ESRF BM29 Martha Brennich (EMBL Grenoble) - - PowerPoint PPT Presentation
Automated data analysis on ESRF BM29 Martha Brennich (EMBL Grenoble) Idealized bio-SAS experiment Solution Scattering Data from Protein of Interest Black Box Neutron source/beamline homesource What can we learn from BioSAXS?
SLIDE 2
SLIDE 3
What can we learn from BioSAXS?
- Low-resolution structural information – shape, overall fold
- Mean molecular weight, oligomeric state
- Mixing ratios
- Model validation
- Domain placement
- Complex structures
- Ab-initio models
- …
?
SLIDE 4
- Dedicated solution
scattering beamline
- Optimized for
macromolecules (4kDa -1MDa)
- Many “non-expert”
users, short visits
SLIDE 5
Automated sample Handling
SLIDE 6
x-rays
sample changer
capillary 3 m detector
Inline HPLC
SLIDE 7
sample changer
SLIDE 8
Automated data acquisition
About 3 minutes per buffer/sample/buffer set Actual acquisition rate: 10 frames/minute
SLIDE 9
ISPyB: Prepare your acquisition from anywhere!
ISPyB: Information System for Protein CrystallographY Beamlines
SLIDE 10
Data Processing - EDNA
2 x 10 10 pyFai Select Average 1 Subtract autorg datgnom dammif damaver dammin
SLIDE 11
Data Processing - EDNA
Image Processing
Radial Integration PyFAI Frame merging and Radition damage detection
1D data reduction
Compare buffers to determine the "Best" Subtract "Best" Buffer from protein curve
Curve reduction
Group all protein curves from same construct Compare curves
Curve Analysis
AutoRg DATGNOM DAMMIF
1D curve Protein Curve Idealized curve Indication of quality (similarity of all curves) Ab-initio Models Model independent Parameters
SLIDE 12
ISPYB: Data Analysis Overview
SLIDE 13
ISPYB: 1d Visualisation
SLIDE 14
ISPYB: Model Visualistation
SLIDE 15
x-rays
sample changer Inline HPLC
SLIDE 16
In-Situ HPLC – increase sample monodispersity
UV cell capillary from GPC MAX pump automatic valve column not controlled by beamline mode valve
SLIDE 17
In-situ HPLC – data acquisition
1000 or more single measurements in a dataset
SLIDE 18
PROCESSING FOR HPLC
1000 pyFai Select Average 1 buffer e.g. frame 1-456 544 samples Subtract 544 autorg 544 peak finder 4 autorg datgnom dammif damaver dammin 4
SLIDE 19
AutoMATED PROCESSING FOR HPLC
Image Processing
Radial Integration PyFAI Merge first frames to create buffer
1D data reduction
Subtract buffer Determine invariants
Curve reduction
Find peaks Merge curves in peak
Curve Analysis
AutoRg DATGNOM DAMMIF
1D curves Protein Curves Idealized curves Ab-initio Models Model independent Parameters
SLIDE 20
HPLC: Real Time feedback
Background quality Signal strength Spoiling?
SLIDE 21
ISPYB: HPLC overview
SLIDE 22
- Data processing framework
- Collaboration between ESRF and Diamond
- Mostly used in macromolecular crystallography
- Python 2.7 based
- At BM29 as a TANGO device
- No direct user interaction: At BM29, the users only need
to explicitly provide sample concentrations
EDNA
SLIDE 23
3 local machines for online processing, in principle each can do everything
BM29 Data Analysis Hardware
Primary Processing Bead modelling HPLC processing XEON 2 core, 3 GHz 2 x XEON 4 core, 2.26 GHz XEON 6 core, 3.40 GHz nVidia Quadro 4000, 2 GB memory nVidia GeForce GTX 750 Ti, 2 GB memory nVidia Quadro M2000, 4 GB memory Before 2009 2011 2016
SLIDE 24
- Reject radiation damaged data
- Identify peaks in HPLC mode
Why do we select frames?
q [nm↑−1 ] log[I(q) ] time
SLIDE 25
- Oversampled data, error bars of each data points non-
ideal (correlated, …)
- Correlation Map (CORMAP) test, originally proposed by
Daniel Franke at EMBL Hamburg
- Core idea: If two frames come from “the same” sample,
the difference between should be random!
- Hence the distribution of + and – differences corresponds
to a series of coin tosses
How do we select frames?
SLIDE 26
CORMAP II
Mark F. Schilling The College Mathematics Journal
- Vol. 21, No. 3 (May, 1990), pp. 196-207
- Distribution is recursive for
the number of coin tosses
- The longest run is actually
pretty short!
- e.g. at BM29 with 1043
q-bins in the range between 7 and 14 points
- Available in freesas
SLIDE 27
- Forward scattering and radius of
gyration are useful for identifying concentration effects on the scattering signal
- But the appropriate data range for
the Guinier approximation is sample dependent and a priori unknown
- Score fits in different regions
- Originally used ATSAS version
- Moved to freeSAS implementation
for HPLC
AutoRg
SLIDE 28
Beam center - the BM29 way
X,Y Transmission 3•10-7
SLIDE 29