Revision: 21188
http://sourceforge.net/p/jmol/code/21188
Author: hansonr
Date: 2016-08-09 16:50:24 +0000 (Tue, 09 Aug 2016)
Log Message:
-----------
MMTF tweak to read entityList and to implement decoding types 12-15
Modified Paths:
--------------
trunk/Jmol/src/org/jmol/adapter/readers/cif/CifReader.java
trunk/Jmol/src/org/jmol/adapter/readers/cif/MMCifReader.java
trunk/Jmol/src/org/jmol/adapter/readers/cif/MMTFReader.java
Modified: trunk/Jmol/src/org/jmol/adapter/readers/cif/CifReader.java
===================================================================
--- trunk/Jmol/src/org/jmol/adapter/readers/cif/CifReader.java 2016-08-09
06:58:10 UTC (rev 21187)
+++ trunk/Jmol/src/org/jmol/adapter/readers/cif/CifReader.java 2016-08-09
16:50:24 UTC (rev 21188)
@@ -533,6 +533,7 @@
private String processChemicalInfo(String type) throws Exception {
if (type.equals("name")) {
chemicalName = data = parser.fullTrim(data);
+ appendLoadNote(chemicalName);
if (!data.equals("?"))
asc.setInfo("modelLoadNote", data);
} else if (type.equals("structuralFormula")) {
Modified: trunk/Jmol/src/org/jmol/adapter/readers/cif/MMCifReader.java
===================================================================
--- trunk/Jmol/src/org/jmol/adapter/readers/cif/MMCifReader.java
2016-08-09 06:58:10 UTC (rev 21187)
+++ trunk/Jmol/src/org/jmol/adapter/readers/cif/MMCifReader.java
2016-08-09 16:50:24 UTC (rev 21188)
@@ -666,7 +666,7 @@
while (parser.getData())
if (!isNull(groupName = getField(CHEM_COMP_ID))
&& !isNull(hetName = getField(CHEM_COMP_NAME)))
- addHetero(groupName, hetName, true);
+ addHetero(groupName, hetName, true, true);
return true;
}
@@ -700,12 +700,12 @@
// return true;
// }
- protected void addHetero(String groupName, String hetName, boolean addNote) {
- if (!vwr.getJBR().isHetero(groupName))
+ protected void addHetero(String groupName, String hetName, boolean doCheck,
boolean addNote) {
+ if (doCheck && !vwr.getJBR().isHetero(groupName))
return;
if (htHetero == null)
htHetero = new Hashtable<String, String>();
- if (htHetero.containsKey(groupName))
+ if (doCheck && htHetero.containsKey(groupName))
return;
htHetero.put(groupName, hetName);
if (addNote)
Modified: trunk/Jmol/src/org/jmol/adapter/readers/cif/MMTFReader.java
===================================================================
--- trunk/Jmol/src/org/jmol/adapter/readers/cif/MMTFReader.java 2016-08-09
06:58:10 UTC (rev 21187)
+++ trunk/Jmol/src/org/jmol/adapter/readers/cif/MMTFReader.java 2016-08-09
16:50:24 UTC (rev 21188)
@@ -45,27 +45,35 @@
*
* see https://github.com/rcsb/mmtf/blob/master/spec.md
*
- * /full/ specification as of 2016.4.18 is implemented,including:
+ * full specification Version: v0.2+dev (as of 2016.08.08) is
+ * implemented,including:
*
- * reading atoms, bonds, and DSSR secondary structure
+ * reading atoms, bonds, and DSSP 1.0 secondary structure
*
- * load =1f88.mmtf
- *
+ * load =1f88.mmtf filter "DSSP1"
*
+ * [Note that the filter "DSSP1" is required, since mmtf included DSSP 1.0
+ * calculations, while the standard for Jmol itself is DSSP 2.0. These two
+ * calculations differ in their treating of helix kinks as one (1.0) or two
+ * (2.0) helices.]
+ *
+ *
* reading space groups and unit cells, and using those as per other readers
*
- * load =1crn.mmtf {1 1 1}
+ * load =1crn.mmtf {1 1 1}
*
- * reading bioassemblies (biomolecules) and applying those transformations
+ * reading bioassemblies (biomolecules) and applying all symmetry
+ * transformations
*
- * load =1auy.mmtf FILTER "biomolecule 1;*.CA,*.P"
+ * load =1auy.mmtf FILTER "biomolecule 1;*.CA,*.P"
*
- * reading biomolecules and lattices, and loading course-grained
+ * reading both biomolecules and lattices, and loading course-grained using the
+ * filter "BYCHAIN" or "BYSYMOP"
*
- * load =1auy.mmtf {2 2 1} filter "biomolecule 1;bychain";spacefill 30.0;
color property symop
+ * load =1auy.mmtf {2 2 1} filter "biomolecule 1;bychain";spacefill 30.0; color
+ * property symop
*
- * Note that the filter "DSSP1" indicates that mmtf included DSSP 1.0
calculation should be used.
- * Otherwise Jmol will use DSSP 2.0 itself.
+ * Many thanks to the MMTF team at RCSB for assistance in this implementation.
*
*/
@@ -78,12 +86,14 @@
/**
* standard set up
+ *
* @param fullPath
* @param htParams
* @param reader
*/
@Override
- protected void setup(String fullPath, Map<String, Object> htParams, Object
reader) {
+ protected void setup(String fullPath, Map<String, Object> htParams,
+ Object reader) {
isBinary = true;
isMMCIF = true;
iHaveFractionalCoordinates = false;
@@ -92,38 +102,318 @@
@Override
protected void processBinaryDocument() throws Exception {
+
+ // load xxx.mmtf filter "..." options
+
+ // NODOUBLE -- standard PDB-like structures
+ // DSSP2 -- use Jmol's built-in DSSP 2.0, not file's DSSP 1.0
+
boolean doDoubleBonds = (!isCourseGrained && !checkFilterKey("NODOUBLE"));
- isDSSP1 = checkFilterKey("DSSP1");
+ isDSSP1 = !checkFilterKey("DSSP2");
boolean mmtfImplementsDSSP2 = false; // so far!
applySymmetryToBonds = true;
map = (new MessagePackReader(binaryDoc, true)).readMap();
+ entities = (Object[]) map.get("entityList");
if (Logger.debugging) {
- for (String s: map.keySet())
+ for (String s : map.keySet())
Logger.info(s);
}
asc.setInfo("noAutoBond", Boolean.TRUE);
Logger.info("MMTF version " + map.get("mmtfVersion"));
Logger.info("MMTF Producer " + map.get("mmtfProducer"));
- appendLoadNote((String) map.get("title"));
+ String title = (String) map.get("title");
+ if (title != null)
+ appendLoadNote(title);
String id = (String) map.get("structureId");
- fileAtomCount = ((Integer)map.get("numAtoms")).intValue();
- int nBonds = ((Integer)map.get("numBonds")).intValue();
+ if (id == null)
+ id = "?";
+ fileAtomCount = ((Integer) map.get("numAtoms")).intValue();
+ int nBonds = ((Integer) map.get("numBonds")).intValue();
Logger.info("id atoms bonds " + id + " " + fileAtomCount + " " + nBonds);
- getAtoms(doDoubleBonds);
+ getMMTFAtoms(doDoubleBonds);
if (!isCourseGrained) {
- getBonds(doDoubleBonds);
- if (isDSSP1 || mmtfImplementsDSSP2)
+ int[] bo = (int[]) decode("bondOrderList");
+ int[] bi = (int[]) decode("bondAtomList");
+ addMMTFBonds(bo, bi, 0, doDoubleBonds, true);
+ if (isDSSP1 || mmtfImplementsDSSP2)
getStructure((int[]) decode("secStructList"));
}
- setSymmetry();
- getBioAssembly();
+ setMMTFSymmetry();
+ getMMTFBioAssembly();
setModelPDB(true);
if (Logger.debuggingHigh)
Logger.info(SV.getVariable(map).asString());
}
-/////////////// MessagePack decoding ///////////////
-
+ //////////////////////////////// MMTF-Specific /////////////////////////
+
+ private Map<String, Object> map; // input JSON-like map from MessagePack
binary file
+ private int fileAtomCount;
+ private int opCount = 0;
+ private int[] groupModels;
+ private String[] labelAsymList; // created in getAtoms; used in
getBioAssembly
+ private Atom[] atomMap; // necessary because some atoms may be deleted.
+ private Object[] entities;
+
+
+ // TODO - also consider mapping group indices
+
+ /**
+ * set up all atoms, including bonding, within a group
+ *
+ * @param doMulti
+ * true to add double bonds
+ *
+ * @throws Exception
+ */
+ @SuppressWarnings("unchecked")
+ private void getMMTFAtoms(boolean doMulti) throws Exception {
+
+ // chains
+ int[] chainsPerModel = (int[]) map.get("chainsPerModel");
+ int[] groupsPerChain = (int[]) map.get("groupsPerChain"); // note that
this is label_asym, not auth_asym
+ labelAsymList = (String[]) decode("chainIdList"); // label_asym
+ String[] authAsymList = (String[]) decode("chainNameList"); // Auth_asym
+
+ // groups
+ int[] groupTypeList = (int[]) decode("groupTypeList");
+ int groupCount = groupTypeList.length;
+ groupModels = new int[groupCount];
+ int[] groupIdList = (int[]) decode("groupIdList");
+ Object[] groupList = (Object[]) map.get("groupList");
+ char[] insCodes = (char[]) decode("insCodeList");
+
+ int[] atomId = (int[]) decode("atomIdList");
+ boolean haveSerial = (atomId != null);
+
+ char[] altloc = (char[]) decode("altLocList"); // rldecode32
+ float[] occ = (float[]) decode("occupancyList");
+
+ float[] x = (float[]) decode("xCoordList");//getFloatsSplit("xCoord",
1000f);
+ float[] y = (float[]) decode("yCoordList");
+ float[] z = (float[]) decode("zCoordList");
+ float[] bf = (float[]) decode("bFactorList");
+ int iatom = 0;
+ String[] nameList = (useAuthorChainID ? authAsymList : labelAsymList);
+ int iModel = -1;
+ int iChain = 0;
+ int nChain = 0;
+ int iGroup = 0;
+ int nGroup = 0;
+ int chainpt = 0;
+ int seqNo = 0;
+ String chainID = "";
+ String authAsym = "", labelAsym = "";
+ char insCode = '\0';
+ atomMap = new Atom[fileAtomCount];
+ for (int j = 0; j < groupCount; j++) {
+ int a0 = iatom;
+ if (insCodes != null)
+ insCode = insCodes[j];
+ seqNo = groupIdList[j];
+ if (++iGroup >= nGroup) {
+ chainID = nameList[chainpt];
+ authAsym = authAsymList[chainpt];
+ labelAsym = labelAsymList[chainpt];
+ nGroup = groupsPerChain[chainpt++];
+ iGroup = 0;
+ if (++iChain >= nChain) {
+ groupModels[j] = ++iModel;
+ nChain = chainsPerModel[iModel];
+ iChain = 0;
+ setModelPDB(true);
+ incrementModel(iModel + 1);
+ nAtoms0 = asc.ac;
+ }
+ }
+ Map<String, Object> g = (Map<String, Object>)
groupList[groupTypeList[j]];
+ String group3 = (String) g.get("groupName");
+ if (vwr.getJBR().isHetero(group3)) {
+ // looking for CHEM_COMP_NAME here... "IRON/SULFUR CLUSTER" for SF4 in
1blu
+ String hetName = "" + g.get("chemCompType");
+ if (htHetero == null || !htHetero.containsKey(group3)) {
+
+ // this is not ideal -- see 1a37.mmtf, where PSE is
LSQRQRST(PSE)TPNVHM
+ // (actually that should be phosphoserine, SPE...)
+ // It is still listed as NON-POLYMER even though it is just a
modified AA.
+
+ if (entities != null && hetName.equals("NON-POLYMER"))
+ out: for (int i = entities.length; --i >= 0;) {
+ Map<String, Object> entity = (Map<String, Object>) entities[i];
+ int[] chainList = (int[]) entity.get("chainIndexList");
+ for (int k = chainList.length; --k >= 0;)
+ if (chainList[k] == iChain) {
+ hetName = "a component of the entity \"" +
entity.get("description") + "\"";
+ break out;
+ }
+ }
+ addHetero(group3, hetName, false, true);
+ }
+ }
+ String[] atomNameList = (String[]) g.get("atomNameList");
+ String[] elementList = (String[]) g.get("elementList");
+ int len = atomNameList.length;
+ for (int ia = 0, pt = 0; ia < len; ia++, iatom++) {
+ Atom a = new Atom();
+ if (insCode != 0)
+ a.insertionCode = insCode;
+ setAtomCoordXYZ(a, x[iatom], y[iatom], z[iatom]);
+ a.elementSymbol = elementList[pt];
+ a.atomName = atomNameList[pt++];
+ if (seqNo >= 0)
+ maxSerial = Math.max(maxSerial, a.sequenceNumber = seqNo);
+ a.group3 = group3;
+ setChainID(a, chainID);
+ if (bf != null)
+ a.bfactor = bf[iatom];
+ if (altloc != null)
+ a.altLoc = altloc[iatom];
+ if (occ != null)
+ a.foccupancy = occ[iatom];
+ if (haveSerial)
+ a.atomSerial = atomId[iatom];
+ if (!filterAtom(a, -1) || !processSubclassAtom(a, labelAsym, authAsym))
+ continue;
+ if (haveSerial) {
+ asc.addAtomWithMappedSerialNumber(a);
+ } else {
+ asc.addAtom(a);
+ }
+ atomMap[iatom] = a;
+ ++ac;
+ }
+ if (!isCourseGrained) {
+ int[] bo = (int[]) g.get("bondOrderList");
+ int[] bi = (int[]) g.get("bondAtomList");
+ addMMTFBonds(bo, bi, a0, doMulti, false);
+ }
+ }
+ }
+
+ private void addMMTFBonds(int[] bo, int[] bi, int a0, boolean doMulti,
boolean isInter) {
+ if (bi == null)
+ return;
+ doMulti &= (bo != null);
+ for (int bj = 0, pt = 0, nj = bi.length / 2; bj < nj; bj++) {
+ Atom a1 = atomMap[bi[pt++] + a0];
+ Atom a2 = atomMap[bi[pt++] + a0];
+ if (a1 != null && a2 != null) {
+ Bond bond = new Bond(a1.index, a2.index, doMulti ? bo[bj] : 1);
+ asc.addBond(bond);
+ if (Logger.debugging && isInter) {
+ Logger.info("inter-group bond " + a1.group3 + a1.sequenceNumber +
"." + a1.atomName
+ + " - " + a2.group3 + a2.sequenceNumber + "." + a2.atomName + " "
+ + bond.order);
+ }
+ }
+ }
+ }
+
+ private void setMMTFSymmetry() {
+ setSpaceGroupName((String) map.get("spaceGroup"));
+ float[] o = (float[]) map.get("unitCell");
+ if (o != null)
+ for (int i = 0; i < 6; i++)
+ setUnitCellItem(i, o[i]);
+ }
+
+ @SuppressWarnings("unchecked")
+ private void getMMTFBioAssembly() {
+ Object[] o = (Object[]) map.get("bioAssemblyList");
+ if (o == null)
+ return;
+ if (vBiomolecules == null)
+ vBiomolecules = new Lst<Map<String, Object>>();
+
+ for (int i = o.length; --i >= 0;) {
+ Map<String, Object> info = new Hashtable<String, Object>();
+ vBiomolecules.addLast(info);
+ int iMolecule = i + 1;
+ checkFilterAssembly("" + iMolecule, info);
+ info.put("name", "biomolecule " + iMolecule);
+ info.put("molecule", Integer.valueOf(iMolecule));
+ Lst<String> assemb = new Lst<String>();
+ Lst<String> ops = new Lst<String>();
+ info.put("biomts", new Lst<M4>());
+ info.put("chains", new Lst<String>());
+ info.put("assemblies", assemb);
+ info.put("operators", ops);
+ // need to add NCS here.
+ Map<String, Object> m = (Map<String, Object>) o[i];
+ Object[] tlist = (Object[]) m.get("transformList");
+ SB chlist = new SB();
+
+ for (int j = 0, n = tlist.length; j < n; j++) {
+
+ Map<String, Object> t = (Map<String, Object>) tlist[j];
+
+ // for every transformation...
+
+ chlist.setLength(0);
+
+ // for compatibility with the mmCIF reader, we create
+ // string lists of the chains in the form $A$B$C...
+
+ int[] chainList = (int[]) t.get("chainIndexList");
+ for (int k = 0, kn = chainList.length; k < kn; k++)
+ chlist.append("$").append(labelAsymList[chainList[k]]);
+ assemb.addLast(chlist.append("$").toString());
+
+ // now save the 4x4 matrix transform for this operation
+
+ String id = "" + (++opCount);
+ addMatrix(id, M4.newA16((float[]) t.get("matrix")), false);
+ ops.addLast(id);
+ }
+ }
+ }
+
+ // Code Name
+ // 0* pi helix
+ // 1 bend (ignored)
+ // 2* alpha helix
+ // 3* extended (sheet)
+ // 4* 3-10 helix
+ // 5 bridge (ignored)
+ // 6* turn
+ // 7 coil (ignored)
+ // -1 undefined
+
+ // 1F88: "secStructList":
[7713371173311776617777666177444172222222222222222222222222222222166771222222222222222222262222222222617662222222222222222222222222222222222177111777722222222222222222221222261173333666633337717776666222222222226622222222222226611771777
+ // DSSP (Jmol):
...EE....EE....TT.....TTT...GGG..HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH.TT...HHHHHHHHHHHHHHHHHHHTHHHHHHHHHHT..TTHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH..........HHHHHHHHHHHHHHHHHHH.HHHHT...EEEETTTTEEEE......TTTTHHHHHHHHHHHTTHHHHHHHHHHHHHTT........
+
+ /**
+ * Get and translate the DSSP string from digit format
+ *
+ * @param a
+ * input data
+ */
+ private void getStructure(int[] a) {
+ BS[] bsStructures = new BS[] { new BS(), null, new BS(), new BS(),
+ new BS(), null, new BS() };
+ if (Logger.debugging)
+ Logger.info(PT.toJSON("secStructList", a));
+ int lastGroup = -1;
+ for (int i = 0; i < a.length; i++) {
+ int type = a[i];
+ switch (type) {
+ case 0: // PI
+ case 2: // alpha
+ case 3: // sheet
+ case 4: // 3-10
+ case 6: // turn
+ bsStructures[type].set(i);
+ lastGroup = i;
+ }
+ }
+
+ int n = (isDSSP1 ? asc.iSet : groupModels[lastGroup]);
+ if (lastGroup >= 0)
+ asc.addStructure(new Structure(n, null, null, null, 0, 0, bsStructures));
+ }
+
+ /////////////// MessagePack decoding ///////////////
+
private Object decode(String key) {
byte[] b = (byte[]) map.get(key);
int type = BC.bytesToInt(b, 0, true);
@@ -132,9 +422,9 @@
switch (type) {
case 1:
return getFloats(b, 4, 1);
- case 2:
- case 3:
- case 4:
+ case 2: // 1-byte
+ case 3: // 2-byte
+ case 4: // 4-byte
return getInts(b, 1 << (type - 2));
case 5:
return rldecode32ToStr(b);
@@ -145,12 +435,18 @@
case 8:
return rldecode32Delta(b, n);
case 9:
- return rldecode32f(b, n, param);
+ return rldecodef(b, n, param);
case 10:
- return recursiveIndex16Deltaf(b, n, param);
+ return unpack16Deltaf(b, n, param);
case 11:
return getFloats(b, 2, param);
- }
+ case 12: // two-byte
+ case 13: // one-byte
+ return unpackf(b, 14 - type, n, param);
+ case 14: // two-byte
+ case 15: // one-byte
+ return unpack(b, 16 - type, n);
+ }
return null;
}
@@ -158,19 +454,19 @@
* mmtf type 1 and 11
*
* byte[4] to float32
- *
+ *
* @param b
- * @param n
- * @param divisor
+ * @param n
+ * @param divisor
* @return float[]
- */
+ */
protected float[] getFloats(byte[] b, int n, float divisor) {
if (b == null)
return null;
int len = (b.length - 12) / n;
float[] a = new float[len];
try {
- switch (n) {
+ switch (n) {
case 2:
for (int i = 0, j = 12; i < len; i++, j += 2)
a[i] = BC.bytesToShort(b, j, false) / divisor;
@@ -184,14 +480,15 @@
}
return a;
}
-
+
/**
* mmtf types 2-4
*
* Decode a byte array into a byte, short, or int array.
- *
+ *
* @param b
- * @param nbytes 1 (byte), 2 (int16), or 4 (int32)
+ * @param nbytes
+ * 1 (byte), 2 (int16), or 4 (int32)
* @return int array
*/
protected int[] getInts(byte[] b, int nbytes) {
@@ -215,14 +512,15 @@
}
return a;
}
-
+
/**
- * mmtf type 5
+ * mmtf type 5
*
- * Decode each four bytes as a 1- to 4-character string label
- * where a 0 byte indicates end-of-string.
- *
- * @param b a byte array
+ * Decode each four bytes as a 1- to 4-character string label where a 0 byte
+ * indicates end-of-string.
+ *
+ * @param b
+ * a byte array
* @return String[]
*/
protected String[] rldecode32ToStr(byte[] b) {
@@ -241,16 +539,15 @@
id[i] = sb.toString();
continue;
}
- }
- }
+ }
+ }
return id;
}
/**
- * mmtf type 6
+ * mmtf type 6
*
- * Decode an array of int32 using run-length decoding to one
- * char per int.
+ * Decode an array of int32 using run-length decoding to one char per int.
*
* @param b
* @param n
@@ -260,16 +557,16 @@
if (b == null)
return null;
char[] ret = new char[n];
- for (int i = 0, pt = 2; i < n;) {
- char val = (char) b[(++pt) << 2];
- for (int j = BC.bytesToInt(b, (++pt) << 2, true); --j >= 0;)
- ret[i++] = val;
+ for (int i = 0, pt = 3; i < n;) {
+ char val = (char) b[(pt++) << 2];
+ for (int j = BC.bytesToInt(b, (pt++) << 2, true); --j >= 0;)
+ ret[i++] = val;
}
return ret;
}
/**
- * mmtf type 7
+ * mmtf type 7
*
* Decode an array of int32 using run-length decoding.
*
@@ -281,10 +578,10 @@
if (b == null)
return null;
int[] ret = new int[n];
- for (int i = 0, pt = 2; i < n;) {
- int val = BC.bytesToInt(b, (++pt) << 2, true);
- for (int j = BC.bytesToInt(b, (++pt) << 2, true); --j >= 0;)
- ret[i++] = val;
+ for (int i = 0, pt = 3; i < n;) {
+ int val = BC.bytesToInt(b, (pt++) << 2, true);
+ for (int j = BC.bytesToInt(b, (pt++) << 2, true); --j >= 0;)
+ ret[i++] = val;
}
return ret;
}
@@ -292,8 +589,7 @@
/**
* mmtf type 8
*
- * Decode an array of int32 using run-length decoding
- * of a difference array.
+ * Decode an array of int32 using run-length decoding of a difference array.
*
* @param b
* @param n
@@ -310,47 +606,47 @@
}
return ret;
}
-
+
/**
- * mmtf type 9
+ * mmtf type 9
*
* Decode an array of int32 using run-length decoding.
*
* @param b
* @param n
- * @param divisor
+ * @param divisor
* @return array of floats
*/
- protected float[] rldecode32f(byte[] b, int n, float divisor) {
+ protected float[] rldecodef(byte[] b, int n, float divisor) {
if (b == null)
return null;
float[] ret = new float[n];
- for (int i = 0, pt = 2; i < n;) {
- int val = BC.bytesToInt(b, (++pt) << 2, true);
- for (int j = BC.bytesToInt(b, (++pt) << 2, true); --j >= 0;)
- ret[i++] = val / divisor;
+ for (int i = 0, pt = 3; i < n;) {
+ int val = BC.bytesToInt(b, (pt++) << 2, true);
+ for (int j = BC.bytesToInt(b, (pt++) << 2, true); --j >= 0;)
+ ret[i++] = val / divisor;
}
return ret;
}
/**
*
- * mmtf type 10 (first part)
+ * mmtf type 10
*
* Decode an array of int16 using run-length decoding of a difference array.
*
* @param b
* @param n
- * @param divisor
+ * @param divisor
* @return array of integers
*/
- protected float[] recursiveIndex16Deltaf(byte[] b, int n, float divisor) {
+ protected float[] unpack16Deltaf(byte[] b, int n, float divisor) {
if (b == null)
return null;
float[] ret = new float[n];
for (int i = 0, pt = 6, val = 0, buf = 0; i < n;) {
int diff = BC.bytesToShort(b, (pt++) << 1, true);
- if (diff == 32767 || diff == -32768) {
+ if (diff == Short.MAX_VALUE || diff == Short.MIN_VALUE) {
buf += diff;
} else {
ret[i++] = (val = val + diff + buf) / divisor;
@@ -360,323 +656,154 @@
return ret;
}
-// /**
-// * Decode an array of int16 using run-length decoding
-// * of a difference array.
-// *
-// * @param b
-// * @param n
-// * @param i0
-// * @return array of integers
-// */
-// protected int[] rldecode16Delta(byte[] b, int n, int i0) {
-// if (b == null)
-// return null;
-// int[] ret = new int[n];
-// for (int i = 0, pt = i0 / 2, val = 0; i < n;) {
-// int diff = BC.bytesToShort(b, (pt++) << 1, true);
-// for (int j = BC.bytesToShort(b, (pt++) << 1, true); --j >= 0;)
-// ret[i++] = (val = val + diff);
-// }
-// return ret;
-// }
-
-
-// /**
-// * Do a split delta to a float[] array
-// * @param xyz label "x", "y", "z", or "bFactor"
-// * @param factor for dividing in the end -- 1000f or 100f
-// * @return float[]
-// *
-// */
-// protected float[] getFloatsSplit(String xyz, float factor) {
-// byte[] big = (byte[]) map.get(xyz + "Big");
-// return (big == null ? null : splitDelta(big,
-// (byte[]) map.get(xyz + "Small"), fileAtomCount, factor));
-// }
-
-// /**
-// * Do a split delta to a float[] array
-// *
-// * @param big
-// * [n m n m n m...] where n is a "big delta" and m is a number of
-// * "small deltas
-// * @param small
-// * array containing the small deltas
-// * @param n
-// * the size of the final array
-// * @param factor
-// * to divide the final result by -- 1000f or 100f here
-// * @return float[]
-// */
-// protected float[] splitDelta(byte[] big, byte[] small, int n, float
factor) {
-// float[] ret = new float[n];
-// for (int i = 0, smallpt = 0, val = 0, datapt = 0, len = big.length >> 2;
i < len; i++) {
-// ret[datapt++] = (val = val + BC.bytesToInt(big, i << 2, true)) /
factor;
-// if (++i < len)
-// for (int j = BC.bytesToInt(big, i << 2, true); --j >= 0; smallpt++)
-// ret[datapt++] = (val = val + BC.bytesToShort(small, smallpt << 1,
true))
-// / factor;
-// }
-// return ret;
-// }
-//
-
-//////////////////////////////// MMTF-Specific /////////////////////////
-
- private Map<String, Object> map; // input JSON-like map from MessagePack
binary file
- private int fileAtomCount;
- private int opCount = 0;
- private int[] groupModels;
-
- private String[] labelAsymList; // created in getAtoms; used in
getBioAssembly
- private Atom[] atomMap; // necessary because some atoms may be deleted.
-
- // TODO - also consider mapping group indices
-
/**
- * set up all atoms, including bonding within a group
*
- * @param doMulti
- * true to add double bonds
+ * mmtf type 12 and 13
*
- * @throws Exception
+ * Unpack an array of int8 or int16 to int32 and divide to give a float32.
+ *
+ * untested
+ *
+ * @param b
+ * @param nBytes
+ * @param n
+ * @param divisor
+ * @return array of integers
*/
- @SuppressWarnings("unchecked")
- private void getAtoms(boolean doMulti) throws Exception {
-
- // chains
- int[] chainsPerModel = (int[]) map.get("chainsPerModel");
- int[] groupsPerChain = (int[]) map.get("groupsPerChain"); // note that
this is label_asym, not auth_asym
- labelAsymList = (String[]) decode("chainIdList"); // label_asym
- String[] authAsymList = (String[]) decode("chainNameList"); // Auth_asym
-
- // groups
- int[] groupTypeList = (int[]) decode("groupTypeList");
- int groupCount = groupTypeList.length;
- groupModels = new int[groupCount];
- int[] groupIdList = (int[]) decode("groupIdList");
- Object[] groupList = (Object[]) map.get("groupList");
- char[] insCodes = (char[]) decode("insCodeList");
-
- //o = map.get("entityList");
- //o = map.get("altLabelList");
- // 1crn: [7, 3, 3, 7, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 6, 6, 6, 7,
7, 2, 2, 2, 2, 2, 2, 2, 2, 1, 7, 3, 3, 7, 1, 1, 1, 7, 7, 7, 4, 4, 4, 7, 7]
-
- int[] atomId = (int[]) decode("atomIdList");
- boolean haveSerial = (atomId != null);
-
- char[] altloc = (char[]) decode("altLocList"); // rldecode32
- float[] occ = (float[]) decode("occupancyList");
-
- float[] x = (float[]) decode("xCoordList");//getFloatsSplit("xCoord",
1000f);
- float[] y = (float[]) decode("yCoordList");
- float[] z = (float[]) decode("zCoordList");
- float[] bf = (float[]) decode("bFactorList");
- int iatom = 0;
- String[] nameList = (useAuthorChainID ? authAsymList : labelAsymList);
- int iModel = -1;
- int iChain = 0;
- int nChain = 0;
- int iGroup = 0;
- int nGroup = 0;
- int chainpt = 0;
- int seqNo = 0;
- String chainID = "";
- String authAsym = "", labelAsym = "";
- char insCode = '\0';
- atomMap = new Atom[fileAtomCount];
- for (int j = 0; j < groupCount; j++) {
- int a0 = iatom;
- if (insCodes != null)
- insCode = insCodes[j];
- seqNo = groupIdList[j];
- if (++iGroup >= nGroup) {
- chainID = nameList[chainpt];
- authAsym = authAsymList[chainpt];
- labelAsym = labelAsymList[chainpt];
- nGroup = groupsPerChain[chainpt++];
- iGroup = 0;
- if (++iChain >= nChain) {
- groupModels[j] = ++iModel;
- nChain = chainsPerModel[iModel];
- iChain = 0;
- setModelPDB(true);
- incrementModel(iModel + 1);
-
- nAtoms0 = asc.ac;
+ protected float[] unpackf(byte[] b, int nBytes, int n, float divisor) {
+ if (b == null)
+ return null;
+ float[] ret = new float[n];
+ switch (nBytes) {
+ case 1:
+ for (int i = 0, pt = 12, offset = 0; i < n;) {
+ int val = b[pt++];
+ if (val == Byte.MAX_VALUE || val == Byte.MIN_VALUE) {
+ offset += val;
+ } else {
+ ret[i++] = (val + offset) / divisor;
+ offset = 0;
}
}
- Map<String, Object> g = (Map<String, Object>)
groupList[groupTypeList[j]];
- String group3 = (String) g.get("groupName");
- addHetero(group3, "" + g.get("chemCompType"), true);
- //System.out.println(group3 + " " + g.get("chemCompType"));
- String[] atomNameList = (String[]) g.get("atomNameList");
- String[] elementList = (String[]) g.get("elementList");
- int len = atomNameList.length;
- for (int ia = 0, pt = 0; ia < len; ia++, iatom++) {
- Atom a = new Atom();
- if (insCode != 0)
- a.insertionCode = insCode;
- setAtomCoordXYZ(a, x[iatom], y[iatom], z[iatom]);
- a.elementSymbol = elementList[pt];
- a.atomName = atomNameList[pt++];
- if (seqNo >= 0)
- maxSerial = Math.max(maxSerial, a.sequenceNumber = seqNo);
- a.group3 = group3;
- setChainID(a, chainID);
- if (bf != null)
- a.bfactor = bf[iatom];
- if (altloc != null)
- a.altLoc = altloc[iatom];
- if (occ != null)
- a.foccupancy = occ[iatom];
- if (haveSerial)
- a.atomSerial = atomId[iatom];
- if (!filterAtom(a, -1) || !processSubclassAtom(a, labelAsym, authAsym))
- continue;
- if (haveSerial) {
- asc.addAtomWithMappedSerialNumber(a);
+ break;
+ case 2:
+ for (int i = 0, pt = 6, offset = 0; i < n;) {
+ int val = BC.bytesToShort(b, (pt++) << 1, true);
+ if (val == Short.MAX_VALUE || val == Short.MIN_VALUE) {
+ offset += val;
} else {
- asc.addAtom(a);
+ ret[i++] = (val + offset) / divisor;
+ offset = 0;
}
- atomMap[iatom] = a;
- ++ac;
}
- if (!isCourseGrained) {
- int[] bo = (int[]) g.get("bondOrderList");
- if (bo != null) {
- int[] bi = (int[]) g.get("bondAtomList");
- for (int bj = 0, pt = 0, nj = bo.length; bj < nj; bj++) {
- Atom a1 = atomMap[bi[pt++] + a0];
- Atom a2 = atomMap[bi[pt++] + a0];
- if (a1 != null && a2 != null)
- addBond(new Bond(a1.index, a2.index, doMulti ? bo[bj] : 1),
false);
- }
- }
- }
+ break;
}
+ return ret;
}
- private void addBond(Bond bond, boolean isInter) {
- asc.addBond(bond);
- if (Logger.debugging && isInter) {
- Atom a1 = asc.atoms[bond.atomIndex1];
- Atom a2 = asc.atoms[bond.atomIndex2];
- Logger.info("bond " +a1.group3 + a1.sequenceNumber + "." + a1.atomName
- + " " + a2.group3 + a2.sequenceNumber + "." + a2.atomName + " " +
bond.order);
+ /**
+ *
+ * mmtf type 14 and 15
+ *
+ * Unpack an array of int8 or int16 to int32.
+ *
+ * untested
+ *
+ * @param b
+ * @param nBytes
+ * @param n
+ * @return array of integers
+ */
+ protected int[] unpack(byte[] b, int nBytes, int n) {
+ if (b == null)
+ return null;
+ int[] ret = new int[n];
+ switch (nBytes) {
+ case 1:
+ for (int i = 0, pt = 12, offset = 0; i < n;) {
+ int val = b[pt++];
+ if (val == Byte.MAX_VALUE || val == Byte.MIN_VALUE) {
+ offset += val;
+ } else {
+ ret[i++] = val + offset;
+ offset = 0;
+ }
+ }
+ break;
+ case 2:
+ for (int i = 0, pt = 6, offset = 0; i < n;) {
+ int val = BC.bytesToShort(b, (pt++) << 1, true);
+ if (val == Short.MAX_VALUE || val == Short.MIN_VALUE) {
+ offset += val;
+ } else {
+ ret[i++] = val + offset;
+ offset = 0;
+ }
+ }
+ break;
}
+ return ret;
}
- private void getBonds(boolean doMulti) {
- int[] b = (int[]) decode("bondOrderList");
- int[] bi = (int[]) decode("bondAtomList");
- for (int i = 0, pt = 0, n = b.length; i < n; i++) {
- Atom a1 = atomMap[bi[pt++]];
- Atom a2 = atomMap[bi[pt++]];
- if (a1 != null && a2 != null)
- addBond(new Bond(a1.index, a2.index, doMulti ? b[i] : 1), true);
- }
- }
+ ///**
+ //* Decode an array of int16 using run-length decoding
+ //* of a difference array.
+ //*
+ //* @param b
+ //* @param n
+ //* @param i0
+ //* @return array of integers
+ //*/
+ //protected int[] rldecode16Delta(byte[] b, int n, int i0) {
+ //if (b == null)
+ //return null;
+ //int[] ret = new int[n];
+ //for (int i = 0, pt = i0 / 2, val = 0; i < n;) {
+ //int diff = BC.bytesToShort(b, (pt++) << 1, true);
+ //for (int j = BC.bytesToShort(b, (pt++) << 1, true); --j >= 0;)
+ //ret[i++] = (val = val + diff);
+ //}
+ //return ret;
+ //}
- private void setSymmetry() {
- setSpaceGroupName((String) map.get("spaceGroup"));
- float[] o = (float[]) map.get("unitCell");
- if (o != null)
- for (int i = 0; i < 6; i++)
- setUnitCellItem(i, o[i]);
- }
+ ///**
+ //* Do a split delta to a float[] array
+ //* @param xyz label "x", "y", "z", or "bFactor"
+ //* @param factor for dividing in the end -- 1000f or 100f
+ //* @return float[]
+ //*
+ //*/
+ //protected float[] getFloatsSplit(String xyz, float factor) {
+ //byte[] big = (byte[]) map.get(xyz + "Big");
+ //return (big == null ? null : splitDelta(big,
+ //(byte[]) map.get(xyz + "Small"), fileAtomCount, factor));
+ //}
- @SuppressWarnings("unchecked")
- private void getBioAssembly() {
- Object[] o = (Object[]) map.get("bioAssemblyList");
- if (vBiomolecules == null)
- vBiomolecules = new Lst<Map<String, Object>>();
+ ///**
+ //* Do a split delta to a float[] array
+ //*
+ //* @param big
+ //* [n m n m n m...] where n is a "big delta" and m is a number of
+ //* "small deltas
+ //* @param small
+ //* array containing the small deltas
+ //* @param n
+ //* the size of the final array
+ //* @param factor
+ //* to divide the final result by -- 1000f or 100f here
+ //* @return float[]
+ //*/
+ //protected float[] splitDelta(byte[] big, byte[] small, int n, float
factor) {
+ //float[] ret = new float[n];
+ //for (int i = 0, smallpt = 0, val = 0, datapt = 0, len = big.length >> 2; i
< len; i++) {
+ //ret[datapt++] = (val = val + BC.bytesToInt(big, i << 2, true)) / factor;
+ //if (++i < len)
+ //for (int j = BC.bytesToInt(big, i << 2, true); --j >= 0; smallpt++)
+ //ret[datapt++] = (val = val + BC.bytesToShort(small, smallpt << 1, true))
+ // / factor;
+ //}
+ //return ret;
+ //}
+ //
- for (int i = o.length; --i >= 0;) {
- Map<String, Object> info = new Hashtable<String, Object>();
- vBiomolecules.addLast(info);
- int iMolecule = i + 1;
- checkFilterAssembly("" + iMolecule, info);
- info.put("name", "biomolecule " + iMolecule);
- info.put("molecule", Integer.valueOf(iMolecule));
- Lst<String> assemb = new Lst<String>();
- Lst<String> ops = new Lst<String>();
- info.put("biomts", new Lst<M4>());
- info.put("chains", new Lst<String>());
- info.put("assemblies", assemb);
- info.put("operators", ops);
- // need to add NCS here.
- Map<String, Object> m = (Map<String, Object>) o[i];
- Object[] tlist = (Object[]) m.get("transformList");
- SB chlist = new SB();
-
- for (int j = 0, n = tlist.length; j < n; j++) {
-
- Map<String, Object> t = (Map<String, Object>) tlist[j];
-
- // for every transformation...
-
- chlist.setLength(0);
-
- // for compatibility with the mmCIF reader, we create
- // string lists of the chains in the form $A$B$C...
-
- int[] chainList = (int[]) t.get("chainIndexList");
- for (int k = 0, kn = chainList.length; k < kn; k++)
- chlist.append("$").append(labelAsymList[chainList[k]]);
- assemb.addLast(chlist.append("$").toString());
-
- // now save the 4x4 matrix transform for this operation
-
- String id = "" + (++opCount);
- addMatrix(id, M4.newA16((float[]) t.get("matrix")), false);
- ops.addLast(id);
- }
- }
- }
-
- // Code Name
- // 0* pi helix
- // 1 bend (ignored)
- // 2* alpha helix
- // 3* extended (sheet)
- // 4* 3-10 helix
- // 5 bridge (ignored)
- // 6* turn
- // 7 coil (ignored)
- // -1 undefined
-
- // 1F88: "secStructList":
[7713371173311776617777666177444172222222222222222222222222222222166771222222222222222222262222222222617662222222222222222222222222222222222177111777722222222222222222221222261173333666633337717776666222222222226622222222222226611771777
- // DSSP (Jmol):
...EE....EE....TT.....TTT...GGG..HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH.TT...HHHHHHHHHHHHHHHHHHHTHHHHHHHHHHT..TTHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH..........HHHHHHHHHHHHHHHHHHH.HHHHT...EEEETTTTEEEE......TTTTHHHHHHHHHHHTTHHHHHHHHHHHHHTT........
-
- /**
- * Get and translate the DSSP string from digit format
- *
- * @param a input data
- */
- private void getStructure(int[] a) {
- BS[] bsStructures = new BS[] { new BS(), null, new BS(), new BS(), new
BS(), null, new BS() };
- if (Logger.debugging)
- Logger.info(PT.toJSON("secStructList", a));
- int lastGroup = -1;
- for (int i = 0; i < a.length; i++) {
- int type = a[i];
- switch (type) {
- case 0: // PI
- case 2: // alpha
- case 3: // sheet
- case 4: // 3-10
- case 6: // turn
- bsStructures[type].set(i);
- lastGroup = i;
- }
- }
-
- int n = (isDSSP1 ? asc.iSet : groupModels[lastGroup]);
- if (lastGroup >= 0)
- asc.addStructure(new Structure(n, null, null, null, 0, 0, bsStructures));
- }
-
}
-
This was sent by the SourceForge.net collaborative development platform, the
world's largest Open Source development site.
------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. http://sdm.link/zohodev2dev
_______________________________________________
Jmol-commits mailing list
Jmol-commits@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-commits
