Mod residues mmCIF export, segmentation data model fixes, other fixes

src/apps/structure-info/model.ts

@@ -119,7 +119,7 @@ export function printSequence(model: Model) {
 
 export function printModRes(model: Model) {
     console.log('\nModified Residues\n=============');
-    const map = model.properties.modifiedResidueNameMap;
+    const map = model.properties.modifiedResidues.parentId;
     const { label_comp_id, _rowCount } = model.atomicHierarchy.residues;
     for (let i = 0; i < _rowCount; i++) {
         const comp_id = label_comp_id.value(i);

src/mol-data/int/impl/segmentation.ts

@@ -50,18 +50,18 @@ export function projectValue({ offsets }: Segmentation, set: OrderedSet, value:
     return OrderedSet.findRange(set, OrderedSet.getAt(offsets, idx), OrderedSet.getAt(offsets, idx + 1) - 1);
 }
 
-export class SegmentIterator<T extends number = number> implements Iterator<Segs.Segment<T>> {
+export class SegmentIterator<I extends number = number> implements Iterator<Segs.Segment<I>> {
     private segmentMin = 0;
     private segmentMax = 0;
     private setRange = Interval.Empty;
-    private value: Segs.Segment<T> = { index: 0, start: 0 as T, end: 0 as T };
+    private value: Segs.Segment<I> = { index: 0 as I, start: 0, end: 0 };
 
     hasNext: boolean = false;
 
     move() {
         while (this.hasNext) {
             if (this.updateValue()) {
-                this.value.index = this.segmentMin++;
+                this.value.index = this.segmentMin++ as I;
                 this.hasNext = this.segmentMax >= this.segmentMin && Interval.size(this.setRange) > 0;
                 break;
             } else {
@@ -75,8 +75,8 @@ export class SegmentIterator<T extends number = number> implements Iterator<Segs
         const segmentEnd = this.segments[this.segmentMin + 1];
         // TODO: add optimized version for interval and array?
         const setEnd = OrderedSet.findPredecessorIndexInInterval(this.set, segmentEnd, this.setRange);
-        this.value.start = Interval.start(this.setRange) as T;
-        this.value.end = setEnd as T;
+        this.value.start = Interval.start(this.setRange);
+        this.value.end = setEnd;
         this.setRange = Interval.ofBounds(setEnd, Interval.end(this.setRange));
         return setEnd > this.value.start;
     }
@@ -94,7 +94,7 @@ export class SegmentIterator<T extends number = number> implements Iterator<Segs
         this.hasNext = this.segmentMax >= this.segmentMin;
     }
 
-    setSegment(segment: Segs.Segment<T>) {
+    setSegment(segment: Segs.Segment<number>) {
         this.setRange = Interval.ofBounds(segment.start, segment.end);
         this.updateSegmentRange();
     }

src/mol-data/int/segmentation.ts

@@ -9,7 +9,7 @@ import OrderedSet from './ordered-set'
 import * as Impl from './impl/segmentation'
 
 namespace Segmentation {
-    export interface Segment<T extends number = number, I extends number = number> { index: number, start: T, end: T }
+    export interface Segment<I extends number = number> { index: I, start: number, end: number }
 
     export const create: <T extends number = number, I extends number = number>(segs: ArrayLike<T>) => Segmentation<T, I> = Impl.create as any;
     export const ofOffsets: <T extends number = number, I extends number = number>(offsets: ArrayLike<T>, bounds: Interval) => Segmentation<T, I> = Impl.ofOffsets as any;
@@ -19,7 +19,7 @@ namespace Segmentation {
     export const projectValue: <T extends number = number, I extends number = number>(segs: Segmentation<T, I>, set: OrderedSet<T>, value: T) => Interval = Impl.projectValue as any;
 
     // Segment iterator that mutates a single segment object to mark all the segments.
-    export const transientSegments: <T extends number = number, I extends number = number>(segs: Segmentation<T, I>, set: OrderedSet<T>, segment?: Segment<T>) => Impl.SegmentIterator<T> = Impl.segments as any;
+    export const transientSegments: <T extends number = number, I extends number = number>(segs: Segmentation<T, I>, set: OrderedSet<T>, segment?: Segment) => Impl.SegmentIterator<I> = Impl.segments as any;
 }
 
 interface Segmentation<T extends number = number, I extends number = number> {

src/mol-geo/representation/structure/visual/util/polymer.ts

@@ -4,7 +4,7 @@
  * @author Alexander Rose <alexander.rose@weirdbyte.de>
  */
 
-import { Unit, StructureElement, StructureProperties, ElementIndex } from 'mol-model/structure';
+import { Unit, StructureElement, StructureProperties } from 'mol-model/structure';
 import { Segmentation } from 'mol-data/int';
 import { MoleculeType } from 'mol-model/structure/model/types';
 import Iterator from 'mol-data/iterator';
@@ -50,7 +50,7 @@ function getTraceName(l: StructureElement) {
     return traceName
 }
 
-function setTraceElement(l: StructureElement, residueSegment: Segmentation.Segment<ElementIndex>) {
+function setTraceElement(l: StructureElement, residueSegment: Segmentation.Segment) {
     const elements = l.unit.elements
     l.element = elements[residueSegment.start]
     const traceName = getTraceName(l)
@@ -97,9 +97,9 @@ const enum AtomicPolymerBackboneIteratorState { nextPolymer, firstResidue, nextR
 export class AtomicPolymerBackboneIterator<T extends number = number> implements Iterator<PolymerBackbonePair> {
     private value: PolymerBackbonePair
 
-    private polymerIt: SegmentIterator<ElementIndex>
-    private residueIt: SegmentIterator<ElementIndex>
-    private polymerSegment: Segmentation.Segment<ElementIndex>
+    private polymerIt: SegmentIterator
+    private residueIt: SegmentIterator
+    private polymerSegment: Segmentation.Segment
     private state: AtomicPolymerBackboneIteratorState = AtomicPolymerBackboneIteratorState.nextPolymer
     private pos: SymmetryOperator.CoordinateMapper
 
@@ -161,8 +161,8 @@ const enum CoarsePolymerBackboneIteratorState { nextPolymer, firstElement, nextE
 export class CoarsePolymerBackboneIterator<T extends number = number> implements Iterator<PolymerBackbonePair> {
     private value: PolymerBackbonePair
 
-    private polymerIt: SegmentIterator<ElementIndex>
-    private polymerSegment: Segmentation.Segment<ElementIndex>
+    private polymerIt: SegmentIterator
+    private polymerSegment: Segmentation.Segment
     private state: CoarsePolymerBackboneIteratorState = CoarsePolymerBackboneIteratorState.nextPolymer
     private pos: SymmetryOperator.CoordinateMapper
     private elementIndex: number
@@ -268,8 +268,8 @@ function createPolymerTraceElement (unit: Unit) {
 export class AtomicPolymerTraceIterator<T extends number = number> implements Iterator<PolymerTraceElement> {
     private value: PolymerTraceElement
 
-    private polymerIt: SegmentIterator<ElementIndex>
-    private residueIt: SegmentIterator<ElementIndex>
+    private polymerIt: SegmentIterator
+    private residueIt: SegmentIterator
     // private polymerSegment: Segmentation.Segment<Element>
     // private state: AtomicPolymerTraceIteratorState = AtomicPolymerTraceIteratorState.nextPolymer
     // private pos: SymmetryOperator.CoordinateMapper

src/mol-model/structure/export/categories/modified-residues.ts

@@ -0,0 +1,61 @@
+/**
+ * Copyright (c) 2017-2018 Mol* contributors, licensed under MIT, See LICENSE file for more info.
+ *
+ * @author David Sehnal <david.sehnal@gmail.com>
+ * @author Alexander Rose <alexander.rose@weirdbyte.de>
+ */
+
+import { Segmentation } from 'mol-data/int';
+import { CifWriter } from 'mol-io/writer/cif';
+import { StructureElement, StructureProperties as P, Unit } from '../../../structure';
+import { CifExportContext } from '../mmcif';
+
+import CifField = CifWriter.Field
+import CifCategory = CifWriter.Category
+
+export function _pdbx_struct_mod_residue(ctx: CifExportContext): CifCategory {
+    const residues = getModifiedResidues(ctx);
+    return {
+        data: residues,
+        name: 'pdbx_struct_mod_residue',
+        fields: pdbx_struct_mod_residue_fields,
+        rowCount: residues.length
+    };
+}
+
+const pdbx_struct_mod_residue_fields: CifField<number, StructureElement[]>[] = [
+    CifField.index('id'),
+    CifField.str(`label_comp_id`, (i, xs) => P.residue.label_comp_id(xs[i])),
+    CifField.int(`label_seq_id`, (i, xs) => P.residue.label_seq_id(xs[i])),
+    CifField.str(`pdbx_PDB_ins_code`, (i, xs) => P.residue.pdbx_PDB_ins_code(xs[i])),
+    CifField.str(`label_asym_id`, (i, xs) => P.chain.label_asym_id(xs[i])),
+    CifField.str(`label_entity_id`, (i, xs) => P.chain.label_entity_id(xs[i])),
+    CifField.str(`auth_comp_id`, (i, xs) => P.residue.auth_comp_id(xs[i])),
+    CifField.int(`auth_seq_id`, (i, xs) => P.residue.auth_seq_id(xs[i])),
+    CifField.str(`auth_asym_id`, (i, xs) => P.chain.auth_asym_id(xs[i])),
+    CifField.str<number, StructureElement[]>('parent_comp_id', (i, xs) => xs[i].unit.model.properties.modifiedResidues.parentId.get(P.residue.label_comp_id(xs[i]))!),
+    CifField.str('details', (i, xs) => xs[i].unit.model.properties.modifiedResidues.details.get(P.residue.label_comp_id(xs[i]))!)
+];
+
+function getModifiedResidues({ model, structure }: CifExportContext): StructureElement[] {
+    const map = model.properties.modifiedResidues.parentId;
+    if (!map.size) return [];
+
+    const ret = [];
+    const prop = P.residue.label_comp_id;
+    const loc = StructureElement.create();
+    for (const unit of structure.units) {
+        if (!Unit.isAtomic(unit) || !unit.conformation.operator.isIdentity) continue;
+        const residues = Segmentation.transientSegments(unit.model.atomicHierarchy.residueAtomSegments, unit.elements);
+        loc.unit = unit;
+        while (residues.hasNext) {
+            const seg = residues.move();
+            loc.element = unit.elements[seg.start];
+            const name = prop(loc);
+            if (map.has(name)) {
+                ret[ret.length] = StructureElement.create(loc.unit, loc.element);
+            }
+        }
+    }
+    return ret;
+}

src/mol-model/structure/export/categories/secondary-structure.ts

@@ -9,11 +9,9 @@ import { CifWriter } from 'mol-io/writer/cif';
 import { SecondaryStructure } from '../../model/properties/seconday-structure';
 import { StructureElement, Unit, StructureProperties as P } from '../../structure';
 import { CifExportContext } from '../mmcif';
-
 import CifField = CifWriter.Field
 import CifCategory = CifWriter.Category
 import { Column } from 'mol-data/db';
-import { ElementIndex } from '../../model';
 
 export function _struct_conf(ctx: CifExportContext): CifCategory {
     const elements = findElements(ctx, 'helix');
@@ -42,7 +40,7 @@ function compare_ssr(x: SSElement<SecondaryStructure.Sheet>, y: SSElement<Second
 
 const struct_conf_fields: CifField[] = [
     CifField.str<number, SSElement<SecondaryStructure.Helix>[]>('conf_type_id', (i, data) => data[i].element.type_id),
-    CifField.str<number, SSElement<SecondaryStructure.Helix>[]>('conf_type_id', (i, data, idx) => `${data[i].element.type_id}${idx + 1}`),
+    CifField.str<number, SSElement<SecondaryStructure.Helix>[]>('id', (i, data, idx) => `${data[i].element.type_id}${idx + 1}`),
     ...residueIdFields('beg_', e => e.start),
     ...residueIdFields('end_', e => e.end),
     CifField.str<number, SSElement<SecondaryStructure.Helix>[]>('pdbx_PDB_helix_class', (i, data) => data[i].element.helix_class),
@@ -53,8 +51,8 @@ const struct_conf_fields: CifField[] = [
 ];
 
 const struct_sheet_range_fields: CifField[] = [
-    CifField.index('id'),
     CifField.str<number, SSElement<SecondaryStructure.Sheet>[]>('sheet_id', (i, data) => data[i].element.sheet_id),
+    CifField.index('id'),
     ...residueIdFields('beg_', e => e.start),
     ...residueIdFields('end_', e => e.end),
     CifField.str('symmetry', (i, data) => '', { valueKind: (i, d) => Column.ValueKind.Unknown })
@@ -66,7 +64,7 @@ function residueIdFields(prefix: string, loc: (e: SSElement<any>) => StructureEl
         CifField.int(`${prefix}label_seq_id`, (i, d) => P.residue.label_seq_id(loc(d[i]))),
         CifField.str(`pdbx_${prefix}PDB_ins_code`, (i, d) => P.residue.pdbx_PDB_ins_code(loc(d[i]))),
         CifField.str(`${prefix}label_asym_id`, (i, d) => P.chain.label_asym_id(loc(d[i]))),
-        CifField.str(`${prefix}_entity_id`, (i, d) => P.chain.label_entity_id(loc(d[i]))),
+        CifField.str(`${prefix}label_entity_id`, (i, d) => P.chain.label_entity_id(loc(d[i]))),
         CifField.str(`${prefix}auth_comp_id`, (i, d) => P.residue.auth_comp_id(loc(d[i]))),
         CifField.int(`${prefix}auth_seq_id`, (i, d) => P.residue.auth_seq_id(loc(d[i]))),
         CifField.str(`${prefix}auth_asym_id`, (i, d) => P.chain.auth_asym_id(loc(d[i])))
@@ -92,7 +90,7 @@ function findElements<T extends SecondaryStructure.Element>(ctx: CifExportContex
         const segs = unit.model.atomicHierarchy.residueAtomSegments;
         const residues = Segmentation.transientSegments(segs, unit.elements);
 
-        let current: Segmentation.Segment<ElementIndex>, move = true;
+        let current: Segmentation.Segment, move = true;
         while (residues.hasNext) {
             if (move) current = residues.move();
 

src/mol-model/structure/export/mmcif.ts

@@ -18,6 +18,7 @@ export interface CifExportContext {
 
 import CifCategory = CifWriter.Category
 import { _struct_conf, _struct_sheet_range } from './categories/secondary-structure';
+import { _pdbx_struct_mod_residue } from './categories/modified-residues';
 
 function copy_mmCif_category(name: keyof mmCIF_Schema) {
     return ({ model }: CifExportContext) => {
@@ -57,6 +58,8 @@ const Categories = [
     copy_mmCif_category('chem_comp'),
     copy_mmCif_category('atom_sites'),
 
+    _pdbx_struct_mod_residue,
+
     // Atoms
     _atom_site
 ];

src/mol-model/structure/model/formats/mmcif.ts

@@ -70,17 +70,20 @@ function getNcsOperators(format: mmCIF_Format) {
     }
     return opers;
 }
-function getModifiedResidueNameMap(format: mmCIF_Format) {
+function getModifiedResidueNameMap(format: mmCIF_Format): Model['properties']['modifiedResidues'] {
     const data = format.data.pdbx_struct_mod_residue;
-    const map = new Map<string, string>();
+    const parentId = new Map<string, string>();
+    const details = new Map<string, string>();
     const comp_id = data.label_comp_id.isDefined ? data.label_comp_id : data.auth_comp_id;
-    const parent_id = data.parent_comp_id;
+    const parent_id = data.parent_comp_id, details_data = data.details;
 
     for (let i = 0; i < data._rowCount; i++) {
-        map.set(comp_id.value(i), parent_id.value(i));
+        const id = comp_id.value(i);
+        parentId.set(id, parent_id.value(i));
+        details.set(id, details_data.value(i));
     }
 
-    return map;
+    return { parentId, details };
 }
 
 function getAsymIdSerialMap(format: mmCIF_Format) {
@@ -132,7 +135,7 @@ function createStandardModel(format: mmCIF_Format, atom_site: AtomSite, entities
         ? format.data.entry.id.value(0)
         : format.data._name;
 
-    const modifiedResidueNameMap = getModifiedResidueNameMap(format);
+    const modifiedResidues = getModifiedResidueNameMap(format);
     const asymIdSerialMap = getAsymIdSerialMap(format)
     const chemicalComponentMap = getChemicalComponentMap(format)
 
@@ -143,14 +146,14 @@ function createStandardModel(format: mmCIF_Format, atom_site: AtomSite, entities
         modelNum: atom_site.pdbx_PDB_model_num.value(0),
         entities,
         symmetry: getSymmetry(format),
-        sequence: getSequence(format.data, entities, atomic.hierarchy, modifiedResidueNameMap),
+        sequence: getSequence(format.data, entities, atomic.hierarchy, modifiedResidues.parentId),
         atomicHierarchy: atomic.hierarchy,
         atomicConformation: atomic.conformation,
         coarseHierarchy: coarse.hierarchy,
         coarseConformation: coarse.conformation,
         properties: {
             secondaryStructure: getSecondaryStructureMmCif(format.data, atomic.hierarchy),
-            modifiedResidueNameMap,
+            modifiedResidues,
             asymIdSerialMap,
             chemicalComponentMap
         },
@@ -163,7 +166,7 @@ function createStandardModel(format: mmCIF_Format, atom_site: AtomSite, entities
 function createModelIHM(format: mmCIF_Format, data: IHMData): Model {
     const atomic = getAtomicHierarchyAndConformation(format, data.atom_site, data.entities);
     const coarse = getIHMCoarse(data);
-    const modifiedResidueNameMap = getModifiedResidueNameMap(format);
+    const modifiedResidues = getModifiedResidueNameMap(format);
     const asymIdSerialMap = getAsymIdSerialMap(format)
     const chemicalComponentMap = getChemicalComponentMap(format)
 
@@ -174,14 +177,14 @@ function createModelIHM(format: mmCIF_Format, data: IHMData): Model {
         modelNum: data.model_id,
         entities: data.entities,
         symmetry: getSymmetry(format),
-        sequence: getSequence(format.data, data.entities, atomic.hierarchy, modifiedResidueNameMap),
+        sequence: getSequence(format.data, data.entities, atomic.hierarchy, modifiedResidues.parentId),
         atomicHierarchy: atomic.hierarchy,
         atomicConformation: atomic.conformation,
         coarseHierarchy: coarse.hierarchy,
         coarseConformation: coarse.conformation,
         properties: {
             secondaryStructure: getSecondaryStructureMmCif(format.data, atomic.hierarchy),
-            modifiedResidueNameMap,
+            modifiedResidues,
             asymIdSerialMap,
             chemicalComponentMap
         },

src/mol-model/structure/model/formats/mmcif/bonds/comp.ts

@@ -141,7 +141,7 @@ export namespace ComponentBond {
             loc.unit = unit;
             while (residues.hasNext) {
                 const seg = residues.move();
-                loc.element = seg.start;
+                loc.element = unit.elements[seg.start];
                 names.add(prop(loc));
             }
         }

src/mol-model/structure/model/model.ts

@@ -42,7 +42,10 @@ export interface Model extends Readonly<{
         /** secondary structure provided by the input file */
         readonly secondaryStructure: SecondaryStructure,
         /** maps modified residue name to its parent */
-        readonly modifiedResidueNameMap: Map<string, string>,
+        readonly modifiedResidues: {
+            parentId: Map<string, string>,
+            details: Map<string, string>
+        },
         /** maps asym id to unique serial number */
         readonly asymIdSerialMap: Map<string, number>
         /** maps residue name to `ChemicalComponent` data */

src/mol-model/structure/query/generators.ts

@@ -7,7 +7,7 @@
 import Query from './query'
 import Selection from './selection'
 import { StructureElement, Unit, StructureProperties as P } from '../structure'
-import { OrderedSet, Segmentation } from 'mol-data/int'
+import { Segmentation } from 'mol-data/int'
 import { LinearGroupingBuilder } from './utils/builders';
 
 export const all: Query.Provider = async (s, ctx) => Selection.Singletons(s, s);
@@ -87,20 +87,20 @@ function atomGroupsSegmented({ entityTest, chainTest, residueTest, atomTest }: A
             const residuesIt = Segmentation.transientSegments(unit.model.atomicHierarchy.residueAtomSegments, elements);
             while (chainsIt.hasNext) {
                 const chainSegment = chainsIt.move();
-                l.element = OrderedSet.getAt(elements, chainSegment.start);
+                l.element = elements[chainSegment.start];
                 // test entity and chain
                 if (!entityTest(l) || !chainTest(l)) continue;
 
                 residuesIt.setSegment(chainSegment);
                 while (residuesIt.hasNext) {
                     const residueSegment = residuesIt.move();
-                    l.element = OrderedSet.getAt(elements, residueSegment.start);
+                    l.element = elements[residueSegment.start];
 
                     // test residue
                     if (!residueTest(l)) continue;
 
                     for (let j = residueSegment.start, _j = residueSegment.end; j < _j; j++) {
-                        l.element = OrderedSet.getAt(elements, j);
+                        l.element = elements[j];
                         if (atomTest(l)) {
                             builder.addElement(l.element);
                         }
@@ -134,20 +134,20 @@ function atomGroupsGrouped({ entityTest, chainTest, residueTest, atomTest, group
             const residuesIt = Segmentation.transientSegments(unit.model.atomicHierarchy.residueAtomSegments, elements);
             while (chainsIt.hasNext) {
                 const chainSegment = chainsIt.move();
-                l.element = OrderedSet.getAt(elements, chainSegment.start);
+                l.element = elements[chainSegment.start];
                 // test entity and chain
                 if (!entityTest(l) || !chainTest(l)) continue;
 
                 residuesIt.setSegment(chainSegment);
                 while (residuesIt.hasNext) {
                     const residueSegment = residuesIt.move();
-                    l.element = OrderedSet.getAt(elements, residueSegment.start);
+                    l.element = elements[residueSegment.start];
 
                     // test residue
                     if (!residueTest(l)) continue;
 
                     for (let j = residueSegment.start, _j = residueSegment.end; j < _j; j++) {
-                        l.element = OrderedSet.getAt(elements, j);
+                        l.element = elements[j];
                         if (atomTest(l)) builder.add(groupBy(l), unit.id, l.element);
                     }
                 }