Rongjin, outliers detection and rejection was inspired by this paper R.Read, Acta Cryst. (1999). D55, 1759-1764. and the paper by R.Agarwal's (197?/198?) mentions this too. You can always turn this off using "main.outliers_rejection=False" if you believe it's necessary. It's probably the quality of your data that causes the amount of outliers you quote. Pavel. PS> FYI: I got your email sent off-list and I always get the emails sent to phenixbb, so there is no need to duplicate them. On 12/8/10 11:45 AM, [email protected] wrote:

Hi All

I have a question regarding reflection outliers detection in phenix.refine: in my input reflection file, there are 24432 reflections, but only 22769 of them were used in refinement. I checked the log file and saw 1663 outliers.

(below if from the log) basic_wilson_outliers = 14 extreme_wilson_outliers = 18 beamstop_shadow_outliers = 0 model_based_outliers = 1657 total = 1663

What is the "model_based_outliers"? What causes so many reflections rejected and not used in the refinement?

Thanks

Rongjin

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

Hi Rongjin Model based outliers are observations reflections that disagree disproportionate with the model, where what is disproportionate is determined along the lines summarized below: Calculated amplitudes and estimated values of alpha and beta are used to compute the log-likelihood of the observed amplitude. The method is inspired by Read, Acta Cryst. (1999). D55, 1759-1764. Outliers are rejected on the basis of the assumption that a scaled log likelihood difference 2(log[P(Fobs)]-log[P(Fmode)])/Q" is distributed according to a Chi-square distribution (Q" is equal to the second derivative of the log likelihood function of the mode of the distribution). Note however that these reflections are not thrown out of your dataset. If you do improve your model and it turns out that these reflections are ok anyway, they will be included later on during refinement. Why these particular reflections are flagged as bad is unclear at the moment. Contact me off list for more help and we can send a summary when done. HTH Peter On 8 December 2010 11:45, [email protected] wrote:

Hi All

I have a question regarding reflection outliers detection in phenix.refine: in my input reflection file, there are 24432 reflections, but only 22769 of them were used in refinement. I checked the log file and saw 1663 outliers.

(below if from the log) basic_wilson_outliers    = 14 extreme_wilson_outliers  = 18 beamstop_shadow_outliers = 0 model_based_outliers     = 1657 total                    = 1663

What is the  "model_based_outliers"?  What causes so many reflections rejected and not used in the refinement?

Thanks

Rongjin

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

-- ----------------------------------------------------------------- P.H. Zwart Research Scientist Berkeley Center for Structural Biology Lawrence Berkeley National Laboratories 1 Cyclotron Road, Berkeley, CA-94703, USA Cell: 510 289 9246 BCSB:      http://bcsb.als.lbl.gov PHENIX:   http://www.phenix-online.org SASTBX:  http://sastbx.als.lbl.gov -----------------------------------------------------------------