Read asym_id for chains from mmCIF

Add a new attribute to Chain particles to
store the mmCIF asym_id, which often differs
from the author-provided "chain ID", and read it
from mmCIF files.

modules/atom/include/Chain.h

@@ -66,6 +66,7 @@ class IMPATOMEXPORT Chain : public Hierarchy {
     m->add_attribute(get_sequence_key(), pi, "");
     m->add_attribute(get_sequence_offset_key(), pi, 0);
     m->add_attribute(get_uniprot_accession_key(), pi, "");
+    m->add_attribute(get_label_asym_id_key(), pi, "");
     m->add_attribute(get_chain_type_key(), pi, UnknownChainType.get_index());
     if (!Hierarchy::get_is_setup(m, pi)) {
       Hierarchy::setup_particle(m, pi);
@@ -90,19 +91,32 @@ class IMPATOMEXPORT Chain : public Hierarchy {
            m->get_has_attribute(get_sequence_offset_key(), pi) &&
            m->get_has_attribute(get_uniprot_accession_key(), pi) &&
            m->get_has_attribute(get_chain_type_key(), pi) &&
+           m->get_has_attribute(get_label_asym_id_key(), pi) &&
            Hierarchy::get_is_setup(m, pi);
   }
 
-  //! Return the chain id
+  //! Return the (author-provided) chain id
   std::string get_id() const {
     return get_model()->get_attribute(get_id_key(), get_particle_index());
   }
 
-  //! Set the chain id
+  //! Set the (author-provided) chain id
   void set_id(std::string c) {
     get_model()->set_attribute(get_id_key(), get_particle_index(), c);
   }
 
+  //! Return the mmCIF asym ID
+  std::string get_label_asym_id() const {
+    return get_model()->get_attribute(get_label_asym_id_key(),
+                                      get_particle_index());
+  }
+
+  //! Set the mmCIF asym ID
+  void set_label_asym_id(std::string c) {
+    get_model()->set_attribute(get_label_asym_id_key(),
+                               get_particle_index(), c);
+  }
+
   //! Return the primary sequence (or any empty string)
   std::string get_sequence() const {
     return get_model()->get_attribute(get_sequence_key(), get_particle_index());
@@ -174,9 +188,12 @@ class IMPATOMEXPORT Chain : public Hierarchy {
   }
 
 
-  //! The key used to store the chain
+  //! The key used to store the author-provided chain ID
   static SparseStringKey get_id_key();
 
+  //! The key used to store the mmCIF asym ID
+  static SparseStringKey get_label_asym_id_key();
+
   //! The key used to store the primary sequence
   static SparseStringKey get_sequence_key();
 

modules/atom/src/Chain.cpp

@@ -30,6 +30,11 @@ SparseStringKey Chain::get_id_key() {
   return k;
 }
 
+SparseStringKey Chain::get_label_asym_id_key() {
+  static SparseStringKey k("label_asym_id");
+  return k;
+}
+
 SparseStringKey Chain::get_sequence_key() {
   static SparseStringKey k("sequence");
   return k;

modules/atom/src/mmcif.cpp

@@ -133,13 +133,15 @@ class AtomSiteCategory : public Category {
     return true;
   }
 
-  void get_chain_particle(const std::string &chain) {
+  void get_chain_particle(const std::string &chain,
+                          const std::string &label_asym_id) {
     if (cp_ == nullptr || chain != curr_chain_) {
       curr_chain_ = chain;
       std::pair<Particle *, std::string> root_chain(root_p_, chain);
       // Check if new chain (for this root)
       if (chain_map_.find(root_chain) == chain_map_.end()) {
         cp_ = internal::chain_particle(model_, chain, filename_);
+        Chain(cp_).set_label_asym_id(label_asym_id);
         Hierarchy(root_p_).add_child(Chain(cp_));
         chain_map_[root_chain] = cp_;
       } else {
@@ -177,9 +179,9 @@ class AtomSiteCategory : public Category {
     // Use author-provided chain ID if available
     std::string label_asym_id = chain_.as_str();
     if (strlen(auth_chain_.as_str()) > 0) {
-      get_chain_particle(auth_chain_.as_str());
+      get_chain_particle(auth_chain_.as_str(), label_asym_id);
     } else {
-      get_chain_particle(label_asym_id);
+      get_chain_particle(label_asym_id, label_asym_id);
     }
     std::string auth_seq_id_str = auth_seq_id_.as_str();
     // Check if new residue

modules/atom/test/test_chain.py

@@ -20,12 +20,15 @@ def test_chain(self):
         self.assertEqual(c.get_sequence(), "")
         self.assertEqual(c.get_sequence_offset(), 0)
         self.assertEqual(c.get_uniprot_accession(), "")
+        self.assertEqual(c.get_label_asym_id(), "")
         c.set_sequence("CCY")
         c.set_sequence_offset(10)
         c.set_uniprot_accession("Q13098")
+        c.set_label_asym_id("X")
         self.assertEqual(c.get_sequence(), "CCY")
         self.assertEqual(c.get_sequence_offset(), 10)
         self.assertEqual(c.get_uniprot_accession(), "Q13098")
+        self.assertEqual(c.get_label_asym_id(), "X")
 
     def test_chain_type(self):
         """Test Chain type"""

modules/atom/test/test_mmcif.py

@@ -26,6 +26,12 @@ def test_read_pdb_or_mmcif(self):
         chains = [IMP.atom.Chain(x)
                   for x in IMP.atom.get_by_type(mp, IMP.atom.CHAIN_TYPE)]
         self.assertEqual(len(chains), 3)
+        self.assertEqual(chains[0].get_id(), "")
+        self.assertEqual(chains[0].get_label_asym_id(), "")
+        self.assertEqual(chains[1].get_id(), "X")
+        self.assertEqual(chains[1].get_label_asym_id(), "B")
+        self.assertEqual(chains[2].get_id(), "A")
+        self.assertEqual(chains[2].get_label_asym_id(), "A")
         self.assertEqual(len(m.get_particle_indexes()), 435)
 
     def test_read_pdb_or_mmcif_no_num(self):