From 3659bf9d0749c9b1c75dba3ef28277f01f69abce Mon Sep 17 00:00:00 2001
From: Gerardo Tauriello <gerardo.tauriello@unibas.ch>
Date: Wed, 28 Nov 2018 17:31:47 +0100
Subject: [PATCH] Clarify lDDT vs QS and chain-penalty.

---
 modules/doc/actions.rst | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/modules/doc/actions.rst b/modules/doc/actions.rst
index 9609642ff..ccda8476e 100644
--- a/modules/doc/actions.rst
+++ b/modules/doc/actions.rst
@@ -38,6 +38,12 @@ In summary it performs the following steps:
   :class:`single chain lDDT scores <ost.mol.alg.lDDTScorer>`,
   :attr:`weighted average of single chain lDDT scores <ost.mol.alg.qsscoring.OligoLDDTScorer.weighted_lddt>`,
   :attr:`lDDT score of oligomer <ost.mol.alg.qsscoring.OligoLDDTScorer.oligo_lddt>`).
+  Note that while the QS score is symmetric (same result when swapping reference
+  and model), the lDDT scores are not. Extra atoms in the model for mapped
+  chains have no effect on the score, while extra atoms in the reference reduce
+  the score. For the oligomeric variants (weighted-lDDT & oligo-lDDT), we do
+  :attr:`penalize for extra chains <ost.mol.alg.qsscoring.OligoLDDTScorer.penalize_extra_chains>`
+  in both reference and model.
 
 .. note ::
 
-- 
GitLab