Wow, can I quote this in the response to the next review that I get back where someone complains about the usage of Rsym? Believe me, I have seen it twice!
Cheers, Steve -----Original Message----- From: CCP4 bulletin board [mailto:[email protected]] On Behalf Of Dale Tronrud Sent: Thursday, January 21, 2010 6:12 PM To: [email protected] Subject: Re: [ccp4bb] R-sym and R-merge I doubt there ever was a universal convention. When I was young, Rsym was calculated from the symmetry related reflections on a single film. With precession photography there could be a lot of them, but even the oscillation films often had many mates because the crystal was aligned with a symmetry axes along the spindle direction. Rmerge was the agreement of the measurements from different films from, maybe, multiple crystals. I think the mates in the Rsym calculation were a subset of those in the Rmerge calculation. Since there was an Rsym for each film, the average of all films was reported. With today's data collection techniques I can't for the life of me figure out a reason for having both Rsym and Rmerge. Dale Tronrud Bart Hazes wrote: > For what it's worth, I've been told that Rmerge was used originally, in > the pre-cryo few images per crystal age, to indicate the R-factor for > merging data collected from multiple crystals. Isym is calculated > identical but refers to data collected from a single crystal. > > To be honest I'm not sure if this reflects history or somebody trying to > make something out of nothing but I've been using Rsym ever since. > > Bart > > Phil Evans wrote: >> I've never been able to discover any difference between Rsym & Rmerge >> Phil >> >> On 21 Jan 2010, at 18:34, james09 pruza wrote: >> >> >>> Dear All CCP4bbers, >>> >>> Please help me in finding *R-sym I *(observed), *R-merge I*(obseved), *R-Free >>> Error i*n the data provided below. If I am not wrong, Can anyone suggest me >>> the difference between R-sym and R-merge? >>> >>> Thanks in advance for the help. >>> >>> James* >>> * >>> >>> R-values of internal consistency : >>> ======== -------------------- >>> (Anomalous scaling counts I+ and I- as separate hkl.) >>> >>> R-merge ( sum |Ij-<I>| / sum |<I>| ) = 23.4 % ( 21.57 % for Fsq > 0) >>> where <I>=sum(Ij)/N, j=1...N, >>> calculated from 188629 observations of 60655 /+//-/ reflexions >>> >>> - dito within user-defined resolution: 23.4 % >>> calculated from 188182 observations of 60508 /+//-/ reflexions >>> >>> >>> Sigma-weighted R-merge : >>> ( sum wj*|Ij-<I>| / sum |<I>| ) = 19.1 % >>> where <I>=sum(wj*Ij), j=1...N, >>> with sum(wj)=1, >>> calculated from 133636 observations of 60655 /+//-/ reflexions >>> >>> - dito within user-defined resolution: 19.1 % >>> calculated from 133323 observations of 60508 /+//-/ reflexions >>> >>> >>> Multiplicity of accepted measurements for unique reflexions, and >>> completeness C >>> in equal-volume shells (on average 4161 possible unique reflexions each): >>> Shell- 1* 2* 3* 4* 5-6* 7-8* 9-12* 13-16* >16* total >>> C[%] >>> limits [A] ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> inf. - 30.0 excluded... (but see right below) >>> 0.0 >>> 30.0 - 6.46 221 878 805 1620 597 0 0 0 0 4121 >>> 95.1 >>> 6.46 - 5.13 73 659 726 1850 840 0 0 0 0 4148 >>> 98.9 >>> 5.13 - 4.48 80 687 841 1823 712 0 0 0 0 4143 >>> 99.1 >>> 4.48 - 4.07 72 673 802 1778 772 0 0 0 0 4097 >>> 99.3 >>> 4.07 - 3.78 94 595 735 1862 831 0 0 0 0 4117 >>> 99.4 >>> 3.78 - 3.56 108 563 738 1883 856 0 0 0 0 4148 >>> 99.7 >>> 3.56 - 3.38 81 519 689 1968 886 0 0 0 0 4143 >>> 99.8 >>> 3.38 - 3.23 49 465 731 1976 877 0 0 0 0 4098 >>> 100.0 >>> 3.23 - 3.11 58 485 772 1950 875 0 0 0 0 4140 >>> 100.0 >>> 3.11 - 3.00 39 457 768 1944 832 0 0 0 0 4040 >>> 98.8 >>> - - - - - - ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> * 30.0 - 3.00 875 5981 7607 18654 8078 0 0 0 0 41195 >>> 99.0 >>> ----------- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> Below specified lower-resolution limit : >>> inf. - 30.0 0 1 0 0 0 0 0 0 0 1 >>> Beyond specified high-resolution limit : >>> 3.00 - 2.99 0 37 25 51 9 0 0 0 0 122 >>> >>> Same multiplicities as relative percentages [%] and <average multiplicity> >>> (total measured unique reflexions in each shell = 100.%) >>> 30.0 - 6.46 5.4 21.3 19.5 39.3 14.5 . . . . >>> <3.38> >>> 6.46 - 5.13 1.8 15.9 17.5 44.6 20.3 . . . . >>> <3.67> >>> 5.13 - 4.48 1.9 16.6 20.3 44.0 17.2 . . . . >>> <3.58> >>> 4.48 - 4.07 1.8 16.4 19.6 43.4 18.8 . . . . >>> <3.61> >>> 4.07 - 3.78 2.3 14.5 17.9 45.2 20.2 . . . . >>> <3.67> >>> 3.78 - 3.56 2.6 13.6 17.8 45.4 20.6 . . . . >>> <3.68> >>> 3.56 - 3.38 2.0 12.5 16.6 47.5 21.4 . . . . >>> <3.74> >>> 3.38 - 3.23 1.2 11.3 17.8 48.2 21.4 . . . . >>> <3.78> >>> 3.23 - 3.11 1.4 11.7 18.6 47.1 21.1 . . . . >>> <3.76> >>> 3.11 - 3.00 1.0 11.3 19.0 48.1 20.6 . . . . >>> <3.77> >>> - - - - - - ----- ----- ----- ----- ----- ----- ----- ----- ----- ->>> >>> * 30.0 - 3.00 2.1 14.5 18.5 45.3 19.6 0.0 0.0 0.0 0.0 100. >>> <3.66> >>> ----------- ----- ----- ----- ----- ----- ----- ----- ----- ----- ==== >>> ------ >>> Below specified lower-resolution limit : >>> inf. - 30.0 . 100.0 . . . . . . . >>> <2.00> >>> Beyond specified high-resolution limit : >>> 3.00 - 2.99 . 30.3 20.5 41.8 7.4 . . . . >>> <3.27> >>> >>> >>> Multiplicity of centric measurements >>> Shell- 1* 2* 3* 4* 5-6* 7-8* 9-12* >13* total / poss. >>> C[%] >>> limits [A] ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> 30.0 - 6.46 93 354 32 2 0 0 0 0 481 566 >>> 85.0 >>> 6.46 - 5.13 37 251 28 0 0 0 0 0 316 334 >>> 94.6 >>> 5.13 - 4.48 39 209 16 0 0 0 0 0 264 281 >>> 94.0 >>> 4.48 - 4.07 24 193 17 0 0 0 0 0 234 249 >>> 94.0 >>> 4.07 - 3.78 25 169 21 0 0 0 0 0 215 223 >>> 96.4 >>> 3.78 - 3.56 34 161 18 0 0 0 0 0 213 220 >>> 96.8 >>> 3.56 - 3.38 25 157 17 0 0 0 0 0 199 204 >>> 97.5 >>> 3.38 - 3.23 26 145 21 0 0 0 0 0 192 192 >>> 100.0 >>> 3.23 - 3.11 26 138 23 0 0 0 0 0 187 187 >>> 100.0 >>> 3.11 - 3.00 27 129 19 0 0 0 0 0 175 175 >>> 100.0 >>> - - - - - - ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> * 30.0 - 3.00 356 1906 212 2 0 0 0 0 2476 2631 >>> 94.1 >>> >>> >>> Multiplicity of acentric measurements for Bijvoet /-/ and /+/ separately : >>> Shell- 1* 2* 3* 4* 5-6* 7-8* 9-12* >13* total / poss. >>> C[%] >>> limits [A] ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> 30.0 - 6.46 1530 4572 790 0 0 0 0 0 6892 7534 >>> 91.5 >>> 6.46 - 5.13 1503 4925 1077 0 0 0 0 0 7505 7720 >>> 97.2 >>> 5.13 - 4.48 1786 4813 974 0 0 0 0 0 7573 7796 >>> 97.1 >>> 4.48 - 4.07 1811 4668 1066 0 0 0 0 0 7545 7754 >>> 97.3 >>> 4.07 - 3.78 1686 4685 1209 0 0 0 0 0 7580 7834 >>> 96.8 >>> 3.78 - 3.56 1756 4595 1308 0 0 0 0 0 7659 7882 >>> 97.2 >>> 3.56 - 3.38 1785 4558 1407 0 0 0 0 0 7750 7896 >>> 98.2 >>> 3.38 - 3.23 1784 4527 1425 0 0 0 0 0 7736 7812 >>> 99.0 >>> 3.23 - 3.11 1833 4549 1415 0 0 0 0 0 7797 7908 >>> 98.6 >>> 3.11 - 3.00 1819 4405 1420 0 0 0 0 0 7644 7826 >>> 97.7 >>> - - - - - - ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- >>> ---- >>> * 30.0 - 3.00 17293 46297 12091 0 0 0 0 0 75681 77962 >>> 97.1 >>> >>> >>> Anomalous R-value : >>> ( sum |<I+>-<I->| / sum |<I+>+<I->| ) = 17.5 % >>> where <I>=sum(Ij)/N, >>> calculated from 37074 anomalous pairs (with 143188 measurements: >>> multiplicity 2.1 <-> 78993 measurements >>> for I+ >>> multiplicity 1.7 <-> 64195 measurements >>> for I- >>> Weighted anomalous R-value : >>> ( sum |<I+>-<I->| / sum |<I+>+<I->| ) = 18.4 % >>> where <I>=sum(wj*Ij) with sum(wj)=1, >>> calculated from 37074 anomalous pairs >>> >>> >>> Table of s-dependent anomalous, and centric, R-factors in equal-volume >>> shells : >>> R_ano. = sum { |<I+> - <I->| } / sum { |<I+> + <I->| } >>> R_cen. = linear R_merge for centric reflexions only. >>> centric >>> Resolution s Bijvoet-pairs <I> R_ano. measurements >>> R_cen. >>> -----[A]---- ------------- ------- --[%]-- | ------- >>> --[%]-- >>> 30.0 - 6.46 0.000 - 0.077 3252 10168. 12.79 | 806 >>> 29.64 >>> 6.46 - 5.13 0.077 - 0.097 3673 2734. 15.19 | 564 >>> 20.88 >>> 5.13 - 4.48 0.097 - 0.112 3694 4719. 16.01 | 440 >>> 24.88 >>> 4.48 - 4.07 0.112 - 0.123 3682 4326. 17.18 | 418 >>> 23.54 >>> 4.07 - 3.78 0.123 - 0.132 3678 2602. 19.76 | 375 >>> 27.71 >>> 3.78 - 3.56 0.132 - 0.141 3724 1490. 22.87 | 337 >>> 29.09 >>> 3.56 - 3.38 0.141 - 0.148 3806 1032. 26.32 | 317 >>> 34.76 >>> 3.38 - 3.23 0.148 - 0.155 3830 728. 31.24 | 317 >>> 35.64 >>> 3.23 - 3.11 0.155 - 0.161 3844 470. 38.98 | 292 >>> 47.76 >>> 3.11 - 3.00 0.161 - 0.167 3779 358. 48.06 | 250 >>> 59.71 >>> - - - - - - -------- ------- ------- | ------- >>> --[%]-- >>> * 30.0 - 3.00 total : 36962 2745. 17.45% | 4116 >>> 28.09% >>> -------- ------- ------- ------- >>> Below specified lower-resolution limit : >>> inf. - 30.0 0.000 - 0.017 1 110. 19.20 | >>> 0 >>> Beyond specified high-resolution limit : >>> 3.00 - 2.99 0.167 - 0.167 111 340. 44.46 | >>> 0 >>> >>> Summary of neg. <I> : 179 centric reflexions >>> 2886 /I-/ >>> 2594 /I+/ >>> including 0 pairs with *both* mates <=0 >>> >>> >>> Tables of s-dependent R-factors in equal-volume shells : >>> (Single measurements or negative <I> omitted from all sums.) >>> R_linear = sum { |I - <I>| } / sum {I} >>> R_square = SQRT( sum{|I - <I>|**2} / sum{I**2} ) >>> >>> Resolution s measurements I<0 <I> <I/s> R_squ. >>> R_lin. >>> -----[A]---- ------------- -------- ----- ------- ----- --[%]-- >>> --[%]-- >>> 30.0 - 6.46 0.000 - 0.077 12293 119 10081. 5.2 27.28 >>> 18.23 >>> 6.46 - 5.13 0.077 - 0.097 13386 394 2490. 2.7 19.45 >>> 18.50 >>> 5.13 - 4.48 0.097 - 0.112 12787 359 4220. 3.0 21.40 >>> 18.89 >>> 4.48 - 4.07 0.112 - 0.123 12670 452 3638. 2.5 21.79 >>> 21.21 >>> 4.07 - 3.78 0.123 - 0.132 12819 799 2102. 1.6 23.85 >>> 26.60 >>> 3.78 - 3.56 0.132 - 0.141 12820 940 1335. 1.1 26.50 >>> 32.14 >>> 3.56 - 3.38 0.141 - 0.148 12270 1186 958. 0.9 30.64 >>> 37.29 >>> 3.38 - 3.23 0.148 - 0.155 12071 1438 661. 0.6 36.79 >>> 44.73 >>> 3.23 - 3.11 0.155 - 0.161 11613 1619 463. 0.5 43.91 >>> 53.68 >>> 3.11 - 3.00 0.161 - 0.167 10870 1708 376. 0.4 51.01 >>> 61.37 >>> - - - - - - ------ ------ ------- ----- ------- >>> ------- >>> * 30.0 - 3.00 total : 123599 9014 2674. 1.9 26.11% >>> 22.61% >>> ------ ------ >>> ======= >>> 4515 non-pos. <I> with 9771 measurements omitted from this table. >>> >>> Table of <<I>/sigma(<I>)> for mean values on output file: >>> ================ >>> Resolution <I> /0/ /+/ /-/ >>> sig<0 >>> -----[A]---- ------------- ------------- ------------- ------------- >>> ----- >>> 30.0 - 6.46 4121 7.7 481 4.5 3551 5.8 3341 >>> 5.7 0 >>> 6.46 - 5.13 4148 4.6 316 2.5 3797 3.5 3708 >>> 3.2 0 >>> 5.13 - 4.48 4143 5.0 264 2.6 3838 3.8 3735 >>> 3.5 0 >>> 4.48 - 4.07 4097 4.3 234 2.1 3835 3.2 3710 >>> 3.0 0 >>> 4.07 - 3.78 4117 3.0 215 1.7 3863 2.2 3717 >>> 2.1 0 >>> 3.78 - 3.56 4148 2.2 213 1.6 3915 1.6 3744 >>> 1.5 0 >>> 3.56 - 3.38 4143 1.6 199 1.3 3938 1.2 3812 >>> 1.1 0 >>> 3.38 - 3.23 4098 1.2 192 0.9 3896 0.8 3840 >>> 0.8 0 >>> 3.23 - 3.11 4140 0.8 187 0.5 3946 0.6 3851 >>> 0.6 0 >>> 3.11 - 3.00 4040 0.6 175 0.4 3853 0.4 3791 >>> 0.4 0 >>> - - - - - - ------------- ------------- ------------- ------------- >>> ----- >>> * 30.0 - 3.00 41195 3.1 2476 2.2 38432 2.3 37249 >>> 2.1 0 >>> ------------- ------------- ------------- ------------- >>> ----- >>> Below specified lower-resolution limit : >>> inf. - 30.0 1 -1.0 0 ... 1 -0.2 1 >>> -1.0 0 >>> Beyond specified high-resolution limit : >>> 3.00 - 2.99 122 0.6 10 0.6 112 0.4 111 >>> 0.4 0 >>> >>> Multiplicity-independent and sigma-weighted R-factors : >>> PCV = "pooled coefficient of variation" >>> R_rim = "redundancy-independent merging R-factor" = R_meas. >>> R_pim = "precision-indicating merging R-factor" >>> Rw_xxx = sigma-weighted R-factors, where : >>> <Iw> = sum{ w * I } with sum{ w } = 1 >>> .. defined as: >>> PCV = sum{ SQRT[ sum{ |Ij - <Ihkl>|**2 }/(n-1) ] } / sum{ <I> } >>> h j h >>> R_rim = sum{ SQRT[n/(n-1)] * sum{ |I - <Ihkl>| } } / sum{ I } >>> R_pim = sum{ SQRT[1/(n-1)] * sum{ |I - <Ihkl>| } } / sum{ I } ) >>> Rw_rim = sum{ SQRT[n/(n-1)] * sum{ w * |I - <Iw>| } } / sum{<Iw>} >>> Rw_pim = sum{ SQRT[1/(n-1)] * sum{ w * |I - <Iw>| } } / sum{<Iw>} >>> Rw_squ = SQRT[ sum{ w * |I - <Iw>|**2 } / sum{<Iw>**2} ] >>> Rw_lin = sum{ w * |I - <Iw>| } / sum{<Iw>} >>> >>> Resolution /+//-/ PCV R_rim R_pim Rw_rim Rw_pim Rw_squ. >>> Rw_lin. >>> -----[A]---- -------- --[%]-- --[%]-- --[%]-- --[%]-- --[%]-- --[%]-- >>> --[%]-- >>> 30.0 - 6.46 5734 25.53 24.98 16.99 13.73 9.41 11.97 >>> 9.96 >>> 6.46 - 5.13 6143 25.77 25.31 17.18 22.87 15.69 15.44 >>> 16.57 >>> 5.13 - 4.48 5903 26.43 25.94 17.69 21.78 14.98 15.39 >>> 15.75 >>> 4.48 - 4.07 5800 29.69 29.07 19.78 25.73 17.66 18.16 >>> 18.65 >>> 4.07 - 3.78 5810 36.85 36.38 24.69 32.42 22.21 20.15 >>> 23.54 >>> 3.78 - 3.56 5768 44.37 43.66 29.37 40.40 27.53 25.67 >>> 29.44 >>> 3.56 - 3.38 5475 51.15 50.32 33.56 46.70 31.61 30.79 >>> 34.22 >>> 3.38 - 3.23 5375 61.60 60.34 40.23 56.21 38.02 37.98 >>> 41.20 >>> 3.23 - 3.11 5175 73.50 72.31 48.10 66.85 45.19 46.53 >>> 49.03 >>> 3.11 - 3.00 4833 83.97 82.72 55.07 77.06 52.10 57.35 >>> 56.51 >>> --------- ------- ------- ------- ------- ------- ------- >>> ------- >>> * total : 56016 31.39% 30.88% 20.91% 24.72% 16.90% 13.95% >>> 17.97% >>> >> >> > > -- > > ======================================================================== ==== > > Bart Hazes (Associate Professor) > Dept. of Medical Microbiology & Immunology > University of Alberta > 1-15 Medical Sciences Building > Edmonton, Alberta > Canada, T6G 2H7 > phone: 1-780-492-0042 > fax: 1-780-492-7521 > > ======================================================================== ==== > Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates Direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system.
