From 0885e63ee5dbe782c02a1b20064195830f3b5a28 Mon Sep 17 00:00:00 2001
From: Xavier Robin <xavier.robin@unibas.ch>
Date: Fri, 19 Jul 2024 09:05:53 +0200
Subject: [PATCH] Document and log about unread models in mmCIF reader
OST only reads the first model (_atom_site.pdbx_PDB_model_num) of an
mmCIF file. This commit clarifies it in the doc, and adds a warning to
the output if models are ignored.
---
modules/io/doc/mmcif.rst | 3 +++
modules/io/pymod/__init__.py | 7 ++++---
modules/io/src/mol/mmcif_reader.cc | 14 ++++++++++----
modules/io/src/mol/mmcif_reader.hh | 1 +
4 files changed, 18 insertions(+), 7 deletions(-)
diff --git a/modules/io/doc/mmcif.rst b/modules/io/doc/mmcif.rst
index 1c7d17c98..5c1874454 100644
--- a/modules/io/doc/mmcif.rst
+++ b/modules/io/doc/mmcif.rst
@@ -105,6 +105,9 @@ Notes:
It is a known limitation of the mmCIF format to allow ambiguous identifiers
for waters (and ligands to some extent) and so we have to require these
additional identifiers.
+* An mmCIF file can contain several models (``atom.site.pdbx_PDB_model_num``).
+ Only the first model occurring in the mmCIF file is read (regardless of the
+ actual model number). If extra models are ignored, a warning is logged.
Info Classes
diff --git a/modules/io/pymod/__init__.py b/modules/io/pymod/__init__.py
index 66a518def..ea3be23a8 100644
--- a/modules/io/pymod/__init__.py
+++ b/modules/io/pymod/__init__.py
@@ -348,9 +348,10 @@ def LoadCHARMMTraj(crd, dcd_file=None, profile='CHARMM',
def LoadMMCIF(filename, fault_tolerant=None, calpha_only=None,
profile='DEFAULT', remote=False, seqres=False, info=False):
"""
- Load a mmCIF file and return one or more entities. Several options allow to
- customize the exact behaviour of the mmCIF import. For more information on
- these options, see :doc:`profile`.
+ Load an mmCIF file and return the first model as an entity.
+
+ Several options allow to customize the exact behaviour of the mmCIF import.
+ For more information on these options, see :doc:`profile`.
Residues are flagged as ligand if they are not waters nor covered by an
``entity_poly`` record (ie. they are non-polymer entities in
diff --git a/modules/io/src/mol/mmcif_reader.cc b/modules/io/src/mol/mmcif_reader.cc
index fd440812e..5ebb954dc 100644
--- a/modules/io/src/mol/mmcif_reader.cc
+++ b/modules/io/src/mol/mmcif_reader.cc
@@ -478,15 +478,21 @@ void MMCifReader::ParseAndAddAtom(const std::vector<StringRef>& columns)
mol::ResNum res_num(0);
bool valid_res_num = false;
if (indices_[PDBX_PDB_MODEL_NUM] != -1) {
+ int model_id = TryGetInt(columns[indices_[PDBX_PDB_MODEL_NUM]],
+ "atom_site.pdbx_PDB_model_num");
if (has_model_) {
- if (curr_model_ != TryGetInt(columns[indices_[PDBX_PDB_MODEL_NUM]],
- "atom_site.pdbx_PDB_model_num")) {
+ if (curr_model_ != model_id) {
+ if (warned_ignored_model_.find(model_id) == warned_ignored_model_.end()) {
+ LOG_WARNING("Ignorning new model " << model_id <<
+ ". Only model " << curr_model_ << " was read.");
+ warned_ignored_model_.insert(model_id);
+ }
return;
}
} else {
+ LOG_INFO("Reading model " << model_id << ".");
has_model_ = true;
- curr_model_ = TryGetInt(columns[indices_[PDBX_PDB_MODEL_NUM]],
- "atom_site.pdbx_PDB_model_num");
+ curr_model_ = model_id;
}
}
diff --git a/modules/io/src/mol/mmcif_reader.hh b/modules/io/src/mol/mmcif_reader.hh
index 23cfc2f4b..8521c048d 100644
--- a/modules/io/src/mol/mmcif_reader.hh
+++ b/modules/io/src/mol/mmcif_reader.hh
@@ -691,6 +691,7 @@ private:
String subst_res_id_; ///< work around for missing label_seq_id's
bool has_model_; ///< keep track of models through different atom_sites
int curr_model_; ///< if we have pdbx_PDB_model_num, store no.
+ std::set<int> warned_ignored_model_; // keep track of ignored model warnings
std::vector<std::pair<mol::ChainHandle, String> > chain_id_pairs_;
///< chain and label_entity_id
MMCifEntityDescMap entity_desc_map_; ///< stores entity items
--
GitLab