EOM - Error in I(0) values

Linear (OLIGOMER), and non-linear (MIXTURE) analysis, singular value decomposition (SVDPLOT), addition of missing fragments (BUNCH, CORAL), analysis of flexible systems (EOM/RANCH & GAJOE), flexible refinement of high-resolution models (SREFLEX)
Post Reply
Message
Author
mawhite
Active member
Posts: 29
Joined: 2011.08.15 22:18
Location: UTMB, Galveston, TX
Contact:

EOM - Error in I(0) values

#1 Post by mawhite » 2022.06.21 19:38

Hello,

I just found this bug in EOM after generating two distributions pools of the same protein in monomer (P1) and dimer (P2) forms. I noticed that the I(0) values in the P1 and P2 pools were considerably different, by a factor of ~14. I also noticed that there is significant variation in the I(0) values in each distribution, although it is proportional to Rg or Dmax and not the total MW as should be expected. I have calculated the average I(0) for each POOL below.

"eom, ATSAS 3.0.4 (r13469)"

Code: Select all

awk '{if($1~"Curve"){getline;II+=$1;ISI+=($1)**2;NN+=1}}END{A=II/NN;printf "P1 AvgI(0)=%.3e +/- %e (N=%d)\n",A,sqrt(ISI-2*A*II+NN*(A**2))/NN,NN}' junP1Ca05e2.eom
P1 AvgI(0)=3.866e+09 +/- 3.113469e+07 (N=5814)

Code: Select all

awk '{if($1~"Curve"){getline;II+=$1;ISI+=($1)**2;NN+=1}}END{A=II/NN;printf "P2 AvgI(0)=%.3e +/- %e (N=%d)\n",A,sqrt(ISI-2*A*II+NN*(A**2))/NN,NN}' junP2Ca05e2.eom
P2 AvgI(0)=5.347e+10 +/- 8.411340e+08 (N=2011)

So for these two distributions P2_I(0) / P1_I(0)= 14
Which is not the expected result based on MW P2/P1 = 2.

This must have a significant effect upon which EOM models P1 or P2 are selected and their fraction contribution to the selected models, in this case overweighting the apparent fraction of P1 in the distribution.

Do you recommend weighting the Fractions of the P2, P1 distributions by their relative I(0) values to get an accurate estimate of their actual occurrence in solution?

Best regards,
Mark

PS. Now that I have your attention, I will repeat my previous request for a 2-D distribution (Rg vs. Dmax) for Gajoe by showing my recent efforts to produce such plots. The 2-D distribution is limited by the PDBs listed in the Gajoe LogFile, which is for the best ensemble only, while it is easy to extract the 2D Rg-Dmax POOL distribution from the EOM Size_list.txt file.

Code: Select all

cat Size_listP1Ca05e2.txt  |awk '{Ro=50.11-0.85;Rm=95.87;Do=177.48-2.76;Dm=326.70;NN=55; dR=(Rm-Ro)/NN;dD=(Dm-Do)/NN;printf "%d %d %8.5f\n",int(($2-Ro)/dR+0.5),int(($3-Do)/dD+0.5),100}' | awk '{NN=55;X=$1;Y=$2;RgDm[X][Y]+=$3;}END{for(i=0;i<=NN;i++){for(j=0;j<=NN;j++){printf "%8.5f ",RgDm[i][j];}printf "\n"}}' |tee RgDmax_P1Ca05e2.PDB-Pool.mat

Code: Select all

grep "e2.pdb "  GA04[1-9]/curve_1/logFile_*_1.log   GA050/curve_1/logFile_*_1.log    |sed 's/~/ /g' |awk '{Ro=50.11-0.85;Rm=95.87;Do=177.48-2.76;Dm=326.70;NN=55; dR=(Rm-Ro)/NN;dD=(Dm-Do)/NN;printf "%d %d %8.5f\n",int(($4-Ro)/dR+0.5),int(($5-Do)/dD+0.5),$6}'  |awk '{NN=55;X=$1;Y=$2;RgDm[X][Y]+=$3*100;}END{for(i=0;i<=NN;i++){for(j=0;j<=NN;j++){printf "%8.5f ",RgDm[i][j];}printf "\n"}}' |tee RgDmax_041-050.PDB.mat
Image

User avatar
Hayds
Active member
Posts: 107
Joined: 2008.05.21 19:01
Location: EMBL, Hamburg

Re: EOM - Error in I(0) values

#2 Post by Hayds » 2022.06.28 17:33

Hi Mark,

to check this it would be good if you could send me a monomer and dimer model. Is that OK?

Cheers,

Haydyn

Post Reply