Ah, this code was double-counting the scale, because the stored peak intensity should already have the scale in the calculation, so I've removed this second scaling, and hopefully proper scales can be used again. (I did find one spot where the stored peak intensity was not scaled up so I've fixed that as well.)
Wayne
On 25 Mar 2014, at 11:42, Wayne Boucher wrote:
> Hello,
>
> I updated the fitting code a month ago to take the scaling into account, because it should, so if you have that update (and it sounds like you must) then that is a partial explanation, although hopefully the scales should be all the same in your case, otherwise this is a problem (!).
>
> Wayne
>
> On 25 Mar 2014, at 11:06, Ivanhoe Leung wrote:
>
>> Dear John, Simon, Wayne,
>>
>> Yes I loaded Topspin processed data directly into CCPNmr Analysis. I also have problems with pseudo 3D datasets.
>>
>> I found a workaround by manually setting the "Scale" to be the same for all spectra under "Experiment:Spectra" before doing the T1/T2 fitting
>>
>> Thanks
>>
>> Ivan
>>
>> ________________________________________
>> From: CcpNmr software mailing list [[log in to unmask]] on behalf of Skinner, Simon P. (Dr.) [[log in to unmask]]
>> Sent: 25 March 2014 10:57
>> To: [log in to unmask]
>> Subject: Re: Problems with "Follow intensity changes"
>>
>> Dear Ivan,
>>
>> Did you use the procedure detailed in the CCPN tutorial to fit your data? Are peaks picked incorrectly by the algorithm in your view?
>>
>> Regards
>>
>> Simon
>>
>> Dr. Simon P. Skinner
>> Postdoctoral Research Associate
>>
>> Department of Biochemistry, University of Leicester
>> Henry Wellcome Building, Lancaster Road
>> Leicester LE1 9HN
>> United Kingdom
>>
>> Email: [log in to unmask]<mailto:[log in to unmask]>
>>
>>
>>
>>
>> On 25 Mar 2014, at 10:01, John Kirkpatrick <[log in to unmask]<mailto:[log in to unmask]>> wrote:
>>
>> Hi Ivan,
>>
>> Perhaps you have sorted this out in the meantime, but if not - can I ask if you are directly loading Topspin-processed data into Analysis?
>>
>> Cheers
>> John
>>
>> On 22/03/14 22:27, Ivanhoe Leung wrote:
>> Hi all,
>>
>> I am using CCPNmr Analysis 2.3.1 for Windows.
>>
>> I ran a series of T2 spectra with different delay times using the Bruker
>> HSQCT2ETF3GPSI parameter set (i.e. not the pseudo 3D version). I created a
>> 'NMR series' with the appropiate delay times and then I use the 'group and
>> fit peaks' function in 'follow intensity changes' to fit the T2 values.
>> However, CCPNmr Analysis didn't seem to be able to follow the intensity
>> changes in my spectra (I tried using 'Volume' or 'Height' options in
>> 'Intensity Type') and subsequently the T2 values that I get from the fittings
>> are incorrect. It seems to me that CCPNmr Analysis is extracting the peak
>> intensity/height with different threshold from the different spectra and it
>> does not follow 'relative peak intensity/height' in the series - is there a
>> quick way to fix it?
>>
>> PS: When I extract the peak area/height manually from TopSpin (using relative
>> intensity) and do the exponential fitting using OriginPro, the fittings are
>> perfect and gave me the expected T2 values so my dataset is OK.
>>
>> Thanks
>>
>> Ivan
>>
>>
>>
>> Ivanhoe Leung | Postdoctoral Research Assistant Chemistry Research
>> Laboratory Oxford University t: +44 (0) 18 6527 5659 | m: +44 (0) 77 8066
>> 3177 Mansfield Road | Oxford OX1 3TA | United Kingdom
>
|