From 164c62e091886855cb16260803060c6e05a5631b Mon Sep 17 00:00:00 2001
From: David Sehnal <david.sehnal@gmail.com>
Date: Sat, 28 Oct 2017 14:28:37 +0200
Subject: [PATCH] more data model
---
.../collections/_spec/segmentation.spec.ts | 7 ++--
.../collections/integer/impl/segmentation.ts | 12 +++---
.../collections/integer/segmentation.ts | 1 -
src/mol-data/atom.ts | 1 +
src/mol-data/model.ts | 15 ++++++--
.../model/{formats => builders}/mmcif.ts | 38 +++++++++++--------
src/mol-data/model/{interfaces => }/common.ts | 0
src/mol-data/model/common/keys.ts | 5 +++
.../model/{interfaces => }/macromolecule.ts | 0
src/mol-data/structure/base.ts | 15 ++++++++
src/script.ts | 4 +-
11 files changed, 66 insertions(+), 32 deletions(-)
rename src/mol-data/model/{formats => builders}/mmcif.ts (71%)
rename src/mol-data/model/{interfaces => }/common.ts (100%)
create mode 100644 src/mol-data/model/common/keys.ts
rename src/mol-data/model/{interfaces => }/macromolecule.ts (100%)
diff --git a/src/mol-base/collections/_spec/segmentation.spec.ts b/src/mol-base/collections/_spec/segmentation.spec.ts
index 005b0bbe8..b44766e5f 100644
--- a/src/mol-base/collections/_spec/segmentation.spec.ts
+++ b/src/mol-base/collections/_spec/segmentation.spec.ts
@@ -20,11 +20,10 @@ describe('segments', () => {
});
it('map', () => {
- const segs = Segmentation.create([1, 2, 3]);
+ const segs = Segmentation.create([0, 1, 2]);
expect(segs.segmentMap).toEqual(new Int32Array([0, 1]));
- expect(segs.offset).toEqual(1);
- expect(Segmentation.getSegment(segs, 1)).toBe(0);
- expect(Segmentation.getSegment(segs, 2)).toBe(1);
+ expect(Segmentation.getSegment(segs, 0)).toBe(0);
+ expect(Segmentation.getSegment(segs, 1)).toBe(1);
})
it('iteration', () => {
diff --git a/src/mol-base/collections/integer/impl/segmentation.ts b/src/mol-base/collections/integer/impl/segmentation.ts
index e24f4757d..a821db8ae 100644
--- a/src/mol-base/collections/integer/impl/segmentation.ts
+++ b/src/mol-base/collections/integer/impl/segmentation.ts
@@ -10,22 +10,22 @@ import Interval from '../interval'
import SortedArray from '../sorted-array'
import Segs from '../segmentation'
-type Segmentation = { segments: SortedArray, segmentMap: ArrayLike<number>, offset: number, count: number }
+type Segmentation = { segments: SortedArray, segmentMap: ArrayLike<number>, count: number }
export function create(values: ArrayLike<number>): Segmentation {
const segments = SortedArray.ofSortedArray(values);
- const min = SortedArray.min(segments), max = SortedArray.max(segments);
- const segmentMap = new Int32Array(max - min);
+ const max = SortedArray.max(segments);
+ const segmentMap = new Int32Array(max);
for (let i = 0, _i = values.length - 1; i < _i; i++) {
- for (let j = values[i] - min, _j = values[i + 1] - min; j < _j; j++) {
+ for (let j = values[i], _j = values[i + 1]; j < _j; j++) {
segmentMap[j] = i;
}
}
- return { segments, segmentMap, offset: min, count: values.length - 1 };
+ return { segments, segmentMap, count: values.length - 1 };
}
export function count({ count }: Segmentation) { return count; }
-export function getSegment({ segmentMap, offset }: Segmentation, value: number) { return segmentMap[value - offset]; }
+export function getSegment({ segmentMap }: Segmentation, value: number) { return segmentMap[value]; }
export function projectValue({ segments }: Segmentation, set: OrderedSet, value: number): Interval {
const last = OrderedSet.max(segments);
diff --git a/src/mol-base/collections/integer/segmentation.ts b/src/mol-base/collections/integer/segmentation.ts
index 5d1e6c3bb..6ecdbdf27 100644
--- a/src/mol-base/collections/integer/segmentation.ts
+++ b/src/mol-base/collections/integer/segmentation.ts
@@ -24,7 +24,6 @@ namespace Segmentation {
interface Segmentation {
'@type': 'segmentation',
readonly segmentMap: ArrayLike<number>,
- readonly offset: number,
readonly count: number
}
diff --git a/src/mol-data/atom.ts b/src/mol-data/atom.ts
index 30f225fa7..037a78c3c 100644
--- a/src/mol-data/atom.ts
+++ b/src/mol-data/atom.ts
@@ -6,6 +6,7 @@
import Tuple from '../mol-base/collections/integer/tuple'
+/** Atom pointer */
interface Atom { '@type': Tuple['@type'] }
namespace Atom {
diff --git a/src/mol-data/model.ts b/src/mol-data/model.ts
index 14d427b43..54f89ceb9 100644
--- a/src/mol-data/model.ts
+++ b/src/mol-data/model.ts
@@ -5,18 +5,25 @@
*/
import * as Formats from './model/formats'
-import CommonInterface from './model/interfaces/common'
-import MacromoleculeInterface from './model/interfaces/common'
+import CommonInterface from './model/common'
+import MacromoleculeInterface from './model/macromolecule'
import Segmentation from '../mol-base/collections/integer/segmentation'
+/**
+ * Interface to the "source data" of the molecule.
+ *
+ * "Atoms" are integers in the range [0, atomCount).
+ */
interface Model {
- data: Formats.RawData,
+ sourceData: Formats.RawData,
common: CommonInterface,
macromolecule: MacromoleculeInterface
+ atomCount: number,
chains: Segmentation,
- residues: Segmentation
+ residues: Segmentation,
+ entities: Segmentation
}
export default Model
\ No newline at end of file
diff --git a/src/mol-data/model/formats/mmcif.ts b/src/mol-data/model/builders/mmcif.ts
similarity index 71%
rename from src/mol-data/model/formats/mmcif.ts
rename to src/mol-data/model/builders/mmcif.ts
index c1cc176dd..71eff5052 100644
--- a/src/mol-data/model/formats/mmcif.ts
+++ b/src/mol-data/model/builders/mmcif.ts
@@ -21,10 +21,11 @@ function findModelBounds(data: mmCIF, startIndex: number) {
function segment(data: mmCIF, bounds: Interval) {
const start = Interval.start(bounds), end = Interval.end(bounds);
- const residues = [start], chains = [start];
+ const residues = [0], chains = [0], entities = [0];
const { label_entity_id, auth_asym_id, auth_seq_id, pdbx_PDB_ins_code } = data.atom_site;
+ let offset = 1;
for (let i = start + 1; i < end; i++) {
const newEntity = !label_entity_id.areValuesEqual(i - 1, i);
const newChain = newEntity || !auth_asym_id.areValuesEqual(i - 1, i);
@@ -32,31 +33,43 @@ function segment(data: mmCIF, bounds: Interval) {
|| !auth_seq_id.areValuesEqual(i - 1, i)
|| !pdbx_PDB_ins_code.areValuesEqual(i - 1, i);
- if (newResidue) residues[residues.length] = i;
- if (newChain) chains[chains.length] = i;
+ if (newEntity) entities[entities.length] = offset;
+ if (newResidue) residues[residues.length] = offset;
+ if (newChain) chains[chains.length] = offset;
+ offset++;
}
- residues[residues.length] = end;
- chains[chains.length] = end;
+ residues[residues.length] = offset;
+ chains[chains.length] = offset;
+ entities[entities.length] = offset;
- return { residues: Segmentation.create(residues), chains: Segmentation.create(chains) };
+ return {
+ residues: Segmentation.create(residues),
+ chains: Segmentation.create(chains),
+ entities: Segmentation.create(entities)
+ };
}
function createModel(raw: RawData, data: mmCIF, bounds: Interval): Model {
const segments = segment(data, bounds);
return {
- data: raw,
+ sourceData: raw,
common: 0 as any,
macromolecule: 0 as any,
+ atomCount: Interval.size(bounds),
residues: segments.residues,
- chains: segments.chains
+ chains: segments.chains,
+ entities: segments.entities
};
}
-function getModels(data: mmCIF): ArrayLike<Model> {
+function buildModels(data: mmCIF): ArrayLike<Model> {
const raw: RawData = { source: 'mmCIF', data };
const models: Model[] = [];
const atomCount = data.atom_site._rowCount;
+
+ if (atomCount === 0) return models;
+
let modelStart = 0;
while (modelStart < atomCount) {
const bounds = findModelBounds(data, modelStart);
@@ -67,9 +80,4 @@ function getModels(data: mmCIF): ArrayLike<Model> {
return models;
}
-export default getModels;
-
-// function createStructure() {
-
-// }
-
+export default buildModels;
\ No newline at end of file
diff --git a/src/mol-data/model/interfaces/common.ts b/src/mol-data/model/common.ts
similarity index 100%
rename from src/mol-data/model/interfaces/common.ts
rename to src/mol-data/model/common.ts
diff --git a/src/mol-data/model/common/keys.ts b/src/mol-data/model/common/keys.ts
new file mode 100644
index 000000000..1a7b5e35b
--- /dev/null
+++ b/src/mol-data/model/common/keys.ts
@@ -0,0 +1,5 @@
+/**
+ * Copyright (c) 2017 molio contributors, licensed under MIT, See LICENSE file for more info.
+ *
+ * @author David Sehnal <david.sehnal@gmail.com>
+ */
diff --git a/src/mol-data/model/interfaces/macromolecule.ts b/src/mol-data/model/macromolecule.ts
similarity index 100%
rename from src/mol-data/model/interfaces/macromolecule.ts
rename to src/mol-data/model/macromolecule.ts
diff --git a/src/mol-data/structure/base.ts b/src/mol-data/structure/base.ts
index e69de29bb..5a03161e3 100644
--- a/src/mol-data/structure/base.ts
+++ b/src/mol-data/structure/base.ts
@@ -0,0 +1,15 @@
+/**
+ * Copyright (c) 2017 molio contributors, licensed under MIT, See LICENSE file for more info.
+ *
+ * @author David Sehnal <david.sehnal@gmail.com>
+ */
+
+import Model from '../model'
+import Structure from '../structure'
+
+//export const Empty { }
+
+export function ofModel(model: Model): Structure {
+ // TODO: create a unit for each chain in the model
+ return 0 as any;
+}
\ No newline at end of file
diff --git a/src/script.ts b/src/script.ts
index 345e177e6..e05c5e817 100644
--- a/src/script.ts
+++ b/src/script.ts
@@ -17,7 +17,7 @@ import CIF from './mol-io/reader/cif'
import Computation from './mol-base/computation'
-import createModels from './mol-data/model/formats/mmcif'
+import buildModels from './mol-data/model/builders/mmcif'
// import { toTypedFrame as applySchema } from './reader/cif/schema'
import { generateSchema } from './mol-io/reader/cif/schema/utils'
@@ -115,7 +115,7 @@ async function runCIF(input: string | Uint8Array) {
console.log(mmcif.pdbx_struct_oper_list.matrix.value(0));
console.time('createModels');
- const models = createModels(mmcif);
+ const models = buildModels(mmcif);
console.timeEnd('createModels');
console.log(models[0].common);
--
GitLab