Re: [ccp4bb] How the IMEAN is calculated in scalepack2mtz

[Karthikeyan Subramanian]

Thank you all for the information and F estimator write-up.

With regards

Karthik



On 13-03-2017 03:56 AM, Phil Evans wrote:

Sorry Ian, I was attributing this to Kevin
apologies
Phil



On 12 Mar 2017, at 21:56, Ian Tickle  wrote:


Eleanor,

Phil is right about the bias of the weighted mean, here's how I do it using a 
minimum mean-squared error estimator which seems to give more accurate results 
than either the unweighted or the weighted mean:

http://staraniso.globalphasing.org/Optimal-F-estimator.pdf

This is actually about Fmean but the same applies to Imean.

This actually only makes a significant difference to the result if there's a 
big difference between sd(I+) and sd(I-), which is obviously not very common.

Cheers

-- Ian



On 12 March 2017 at 17:58, Eleanor Dodson  wrote:
  You read:
h k l  IPLUS   SIGIPLUS INEG   SIGINEG
Then program calculates this:

SIGIMEAN = SIGIPLUS*SIGINEG/(SIGIPLUS+SIGINEG)

  IMEAN = (IPLUS/SIGIPLUS + INEG/SIGINEG)*SIGIMEAN

ie: IMEAN = ( IPLUS*SIGINEG   + INEG * SIGIPLUS ) / /(SIGIPLUS+SIGINEG)

Is that the right thing to do? Not sure!

Eleanor Dodson



On 11 March 2017 at 15:18, Karthikeyan Subramanian  wrote:
Dear CCP4bb,

How IMEAN and SIGIMEAN is calculated in scalepack2mtz if the input is with 
anomalous intensity (obtained from HKL2000). Any guide/reference is highly 
appreciated.

Thanks in advance,

With regards

Karthikeyan S.

Principal Scientist

IMTECH, Chandigarh

[ccp4bb] PostDoc opportunities at RCSB PDB

[Andreas Prlic]

We are looking for talented and highly motivated Postdoctoral Fellows to join
the multidisciplinary development team at RCSB PDB.

http://www.rcsb.org/pdb/news.do?year=2017=58b73b475ae81f58037bf5a8

The Challenge: Develop innovative analysis, integration, query, and
visualization
tools for 3D biomolecular structures to help accelerate research and
training in biology, medicine, and related disciplines. In these projects,
we employ the latest advances in computer science to develop highly
interactive features and scalable services and workflows. The overall goal
is always to develop highly interactive features for the RCSB PDB website.

Two positions are available: one focused on Visualization (position #1104) and
Machine Learning/Biological Assemblies (position #1105). Possible project
areas include:

- Innovative representations for 3D structural data, including biological
assemblies
- Visualizations for comparative analyses
- Multiscale rendering for exploring large structures
- Using machine learning to build the next generation search engine
for molecular
biology

This is a unique opportunity to engage in leading edge research, development,
and outreach activities of the RCSB PDB with worldwide impact.

--
---
Dr. Andreas Prlic
RCSB PDB Protein Data Bank
Technical & Scientific Team Lead
University of California, San Diego

Editor Software Section
PLOS Computational Biology

BioJava Project Lead
---

[ccp4bb] Postdoc positions at Uppsala University

[Maria Selmer]

Dear all,
We have openings for two postdoctoral researchers for a project on protein 
synthesis and antibiotic resistance at Uppsala University, Sweden. 
The postdocs could also partly be involved in our ongoing studies of protein 
evolution.

On-line advert with more information and instructions for on-line application 
can be found here: 
http://www.uu.se/en/about-uu/join-us/details/?positionId=137102 
http://www.uu.se/en/about-uu/join-us/details/?positionId=137081

For more information, contact maria.sel...@icm.uu.se

Application deadline: March 19.
/Maria


Prof. Maria Selmer  
Deputy Head of Department
Cell and Molecular Biology, Uppsala University
BMC, Box 596
SE-75124 Uppsala
Sweden

Telephone: +46-(0)18-4714177
Fax: +46-(0)18-536971

[ccp4bb] surface area

[chemocev marker]

Hi
I am comparing some protein complex with PISA analysis. Can someone make a
note what is difference between interface area and buried surface area. I
think buried surface area also include the interface area + the area
encloses with in the protein. Does it make sense to mentioned the interface
area separately if we count the buried area??

best

Jiri

Re: [ccp4bb] ADD sulfinic-cysteine to protein aa-chain

[Renato Mateus Domingos]

Thanks all for the help!

Definitely, using: Extensions -> Modelling -> Replace Residue ... -> CSD ->
Mutate was the easiest way to solve the problem of adding the sulfinic
cysteine!

Thanks
Renato

*Renato Mateus Domingos*
*PhD student at Departamento de Genética e Biologia Evolutiva*
*Universidade de São Paulo, Brazil*
*Phone: (+55)(11)30917590*

On 12 March 2017 at 16:31, Paul Emsley  wrote:

> On 11/03/2017 17:37, Renato Mateus Domingos wrote:
>
>> Hi
>>
>
> Hi.
>
> I add the sulfinic cysteine in my coordinate map, however I am strugling
>> to make proper
>> bonds between the aa chain of the protein and the sulfinic cysteine I
>> added. the PDB does
>> not recognize the modified cysteine I added as an amino acid. Any help
>> please?
>>
>
> This is not a very clear question (although my Portuguese is not greatest
> :))
>
> You mean SH -> S(=O)O[H] ?
>
> If so, I think that you want to replace your CYS with a CSD
>
> If you're using Coot:
>
> 1) centre on the CYS of interest
>
> 2 ) Extensions -> Modelling -> Replace Residue ... -> CSD -> Mutate
>
>
> Paul.
>
>

Re: [ccp4bb] How the IMEAN is calculated in scalepack2mtz

[Ian Tickle]

Hi Eleanor

I reran my simulation including your method.  Let's just look at the two
extreme cases in my original table: sd(F+) = 10, sd(F-) = 1 (5th row) and
sd(F+) = 1000, sd(F-) = 100 (last row: the case where my method performs
worst but still better than the others!).  I used the same 'true' values as
before: F+ = 578, F- = 1369, Fmean = 973.

For the first case, the values of su(Fmean) for the methods 'unweighted',
'variance-weighted', 'RMS-error-weighted' (my method) and 'sd-weighted'
(your method) are 5 1 5 1 respectively, so as expected the two weighted
methods give the lowest variances.

Similarly the mean biases are: 0 388 0 323, so again as expected the
weighted methods are the most biased.

The RMS errors are: 5 388 5 323 so in this case the 'variance-weighted' and
'sd-weighted' methods both perform poorly with the highest overall errors,
though your method is somewhat better than the 'variance-weighted' one.

For each statistic I'm including the first 3 values for comparison but they
are of course exactly the same as before: only the last value for the
'sd-weighted' method is new.

For the second case with sd(F+) = 1000, sd(F-) = 100 and the same values
for the true Fs, the results from the 4 methods in the same order as before
are:

su(Fmean):38699  171  115.

bias(Fmean):87  389  274  339.

RMSE(Fmean):   396  402  323  358.

So again the results for su(Fmean) and bias(Fmean) are as expected (i.e.
the 'sd-weighted' method performs intermediate between 'unweighted' and
'variance-weighted').  However using the RMSE measure the
'RMS-error-weighted' method performs best.  Using the RMSE measure as the
definitive metric (which IMO is the only sensible procedure), in no case
does the 'RMS-error-weighted' method perform worse than any of the others.

Cheers

-- Ian


On 13 March 2017 at 08:22, Ian Tickle  wrote:

>
> Eleanor
>
> I notice that you are calculating a weighted mean using 1/sd(I) as the
> weight whereas in least squares one would of course normally use
> variance-weighting, i.e. 1/sd(I)^2.  I assume this is intentional and is a
> way to reduce the bias effect of weighting, so that the resulting bias of
> the mean would be intermediate between that of an unweighted and a
> variance-weighted mean, though the variance of the mean would no longer be
> a minimum.
>
> Assuming that it was indeed intentional, I will see if I can find the
> program I used to simulate the various ways of calculating the mean using
> the MSE as the measure of accuracy and see how this method compares with
> the others.
>
> Cheers
>
> -- Ian
>
>
>
> On 12 March 2017 at 17:58, Eleanor Dodson 
> wrote:
>
>>  You read:
>> h k l  IPLUS   SIGIPLUS INEG   SIGINEG
>> Then program calculates this:
>>
>>SIGIMEAN = SIGIPLUS*SIGINEG/(SIGIPLUS+SIGINEG)
>>
>>  IMEAN = (IPLUS/SIGIPLUS + INEG/SIGINEG)*SIGIMEAN
>>
>> ie: IMEAN = ( IPLUS*SIGINEG   + INEG * SIGIPLUS ) / /(SIGIPLUS+SIGINEG)
>>
>> Is that the right thing to do? Not sure!
>>
>> Eleanor Dodson
>>
>>
>>
>> On 11 March 2017 at 15:18, Karthikeyan Subramanian 
>> wrote:
>>
>>> Dear CCP4bb,
>>>
>>> How IMEAN and SIGIMEAN is calculated in scalepack2mtz if the input is
>>> with anomalous intensity (obtained from HKL2000). Any guide/reference is
>>> highly appreciated.
>>>
>>> Thanks in advance,
>>>
>>> With regards
>>>
>>> Karthikeyan S.
>>>
>>> Principal Scientist
>>>
>>> IMTECH, Chandigarh
>>>
>>
>>
>

[ccp4bb] Vacant positions as PhD student in Medical Structural Biology

[Karin Lindkvist]

Dear colleagues and structural biologist,

I want to bring your attention to that we have two vacant positions as PhD 
students in Medical Structural Biology at Lund University, Sweden, available. 
The last application date is 31st of March.


https://lu.mynetworkglobal.com/se/what:job/jobID:139767/iframeEmbedded:0/where:4

https://lu.mynetworkglobal.com/se/what:job/jobID:139773/iframeEmbedded:0/where:4


With best wishes,
Karin Lindkvist


***
Dr. Karin Lindkvist
Associate Professor
Experimental Medical Science
Medical Structural Biology
Lund University
BMC C13
S-221 84 Lund
Sweden

Tel: +46 2228041
Tel (mobile): +46 734222786
Email: karin.lindkv...@med.lu.se

Re: [ccp4bb] How the IMEAN is calculated in scalepack2mtz

[Ian Tickle]

Eleanor

I notice that you are calculating a weighted mean using 1/sd(I) as the
weight whereas in least squares one would of course normally use
variance-weighting, i.e. 1/sd(I)^2.  I assume this is intentional and is a
way to reduce the bias effect of weighting, so that the resulting bias of
the mean would be intermediate between that of an unweighted and a
variance-weighted mean, though the variance of the mean would no longer be
a minimum.

Assuming that it was indeed intentional, I will see if I can find the
program I used to simulate the various ways of calculating the mean using
the MSE as the measure of accuracy and see how this method compares with
the others.

Cheers

-- Ian



On 12 March 2017 at 17:58, Eleanor Dodson  wrote:

>  You read:
> h k l  IPLUS   SIGIPLUS INEG   SIGINEG
> Then program calculates this:
>
>SIGIMEAN = SIGIPLUS*SIGINEG/(SIGIPLUS+SIGINEG)
>
>  IMEAN = (IPLUS/SIGIPLUS + INEG/SIGINEG)*SIGIMEAN
>
> ie: IMEAN = ( IPLUS*SIGINEG   + INEG * SIGIPLUS ) / /(SIGIPLUS+SIGINEG)
>
> Is that the right thing to do? Not sure!
>
> Eleanor Dodson
>
>
>
> On 11 March 2017 at 15:18, Karthikeyan Subramanian 
> wrote:
>
>> Dear CCP4bb,
>>
>> How IMEAN and SIGIMEAN is calculated in scalepack2mtz if the input is
>> with anomalous intensity (obtained from HKL2000). Any guide/reference is
>> highly appreciated.
>>
>> Thanks in advance,
>>
>> With regards
>>
>> Karthikeyan S.
>>
>> Principal Scientist
>>
>> IMTECH, Chandigarh
>>
>
>