luancarvalhomartins / PyAutoFEP

PyAutoFEP: an automated FEP workflow for GROMACS integrating enhanced sampling methods
163 stars 76 forks source link

Mismatch Results #108

Closed xy21hb closed 1 year ago

xy21hb commented 1 year ago

Dear all,

I rerun the tutorial for a second time with exactly the same parameters, but the results look so different, any idea?

FXR_12→FXR_85 17.2±0.8 kcal/mol 17.8±0.9 kcal/mol ΔΔG = -0.6±0.9 kcal/mol FXR_12→FXR_84 -2.0±0.6 kcal/mol -2.7±0.7 kcal/mol ΔΔG = 0.8±0.6 kcal/mol FXR_12→FXR_88 0.7±378.4 kcal/mol -6.6±1.1 kcal/mol ΔΔG = 7.3±20.5 kcal/mol FXR_12→FXR_76 -5.8±2.7 kcal/mol -6.1±0.9 kcal/mol ΔΔG = 0.3±1.6 kcal/mol FXR_12→FXR_74 1.0±0.2 kcal/mol 0.9±0.2 kcal/mol ΔΔG = 0.1±0.2 kcal/mol ================= ΔΔG to FXR_12 ================== Molecule ΔΔG
FXR_12 0.0 kcal/mol FXR_74 0.1 kcal/mol FXR_76 0.3 kcal/mol FXR_84 0.8 kcal/mol FXR_85 -0.6 kcal/mol FXR_88 7.3 kcal/mol

======================================= FXR_12→FXR_74 1.0±0.0 kcal/mol 1.0±0.0 kcal/mol ΔΔG = 0.0±0.0 kcal/mol FXR_12→FXR_88 -5.8±0.3 kcal/mol -5.9±0.3 kcal/mol ΔΔG = 0.2±0.3 kcal/mol FXR_12→FXR_76 -5.7±0.1 kcal/mol -5.3±0.1 kcal/mol ΔΔG = -0.3±0.1 kcal/mol FXR_12→FXR_84 -3.2±0.1 kcal/mol -2.7±0.1 kcal/mol ΔΔG = -0.6±0.1 kcal/mol FXR_12→FXR_85 17.3±0.1 kcal/mol 16.7±0.1 kcal/mol ΔΔG = 0.6±0.1 kcal/mol ================= ΔΔG to FXR_12 ================== Molecule ΔΔG
FXR_12 0.0 kcal/mol FXR_74 0.0 kcal/mol FXR_76 -0.3 kcal/mol FXR_84 -0.6 kcal/mol FXR_85 0.6 kcal/mol FXR_88 0.2 kcal/mol

luancarvalhomartins commented 1 year ago

That's expected behavior not related to PyAutoFEP itself, but to the underlying theory of MD sampling. The results will always vary. For some discussion on the statistics of the results, see: