[RASMB] on low s value peak in sedfit

[John Philo]

Chin,
 
First, the proper statistical way to address any issue about whether a
change in rms residual is significant is to apply an F-test.
 
While I would guess that you are indeed detecting sedimentation of a real
low s species, I think the question you should be asking yourself is whether
the area of that peak has any relevance for what you are trying to do. That
is, this material cannot possibly be your protein or a protein
aggregate---most likely you are detecting sedimentation of some buffer
component. 
 
If this peak isn't due to protein, it shouldn't be counted in calculating
fraction main peak or any other property of the protein (for example
weight-average sedimentation coefficient), so why do you care whether its
area is accurate? Just copy the distribution table to another program and
manually delete the low s value region, or don't include it when you ask
SEDFIT to integrate the areas.
 
With regard to the half-peak at s-max, that is quite different. The presence
of that half-peak probably indicates the presence of material sedimenting
faster than s-max, and there is no guarantee that the area of that half-peak
accurately measures the amount of such material. Thus if you want to
characterize all the aggregates in your sample  then in my opinion you
really must raise s-max until you no longer see a half-peak at the upper
limit.
 
John Philo
Alliance Protein Laboratories

-----Original Message-----
From: rasmb-admin at server1.bbri.org [mailto:rasmb-admin at server1.bbri.org] On
Behalf Of Qin Zou
Sent: Thursday, October 28, 2004 5:41 AM
To: rasmb at server1.bbri.org
Subject: [RASMB] on low s value peak in sedfit



Dear experts, 
I have been reading some discussions on low s value peaks using sedfit in
this group. I have such a partial peak with smin at 0.1s. I have tried Dr.
Schuck's method that is to increase s-min value gradually while monitoring
the rmsd. Here is some of the results: 
s-min: 0.2s        0.5s        1.0s                1.2s                1.4s

rmsd:  0.005249        0.0053        0.005556        0.005580
0.005588 

The height of the partial peak did drop but never completely disappears.
Couple of questions: 
1)Is the increase of the rmsd shown above considered significant? 
2)If there is a sedimenting component at this low s, how do I integrate this
partial peak and the other complete peaks? It seems that moving around s-min
affects the integration alot. 
3)There is also a little partial peak at s-max. If I only want to know the
loading percentage of all the aggregates, should I just integrate all the
way to the s-max? 

Thanks in advance. 
Regards, 

Chin  Zou

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20041101/320e5f08/attachment.htm>