String codons() {
char seq[] = chromoSequence.toCharArray();
char seqNew[] = chromoNewSequence.toCharArray();
codonsOld = "";
codonsNew = "";
int codonIdx = 0;
int i = 0;
int step = transcript.isStrandPlus() ? 1 : -1;
char codonOld[] = new char[3];
char codonNew[] = new char[3];
for (Exon ex : transcript.sortedStrand()) {
int start = ex.isStrandPlus() ? ex.getStart() : ex.getEnd();
for (i = start; ex.intersects(i); i += step, codonIdx = (codonIdx + 1) % 3) {
codonOld[codonIdx] = seq[i];
codonNew[codonIdx] = seqNew[i];
if (codonIdx == 2) addIfDiff(codonOld, codonNew);
}
}
for (; codonIdx != 0; i += step, codonIdx = (codonIdx + 1) % 3) {
codonOld[codonIdx] = 'N';
codonNew[codonIdx] = 'N';
if (codonIdx == 2) addIfDiff(codonOld, codonNew);
}
return codonsOld + "/" + codonsNew;
}
/**
* Find the last position where a nonsense mediated decay is supposed to occurr This is 50 bases
* (MND_BASES_BEFORE_LAST_JUNCTION bases) before the last exon-exon junction.
*
* @param tr
* @return
*/
public int lastNmdPos(Transcript tr) {
// ---
// Get last exon
// ---
int cdsEnd = tr.getCdsEnd();
int cdsStart = tr.getCdsStart();
Marker cds =
new Marker(
tr.getChromosome(),
Math.min(cdsStart, cdsEnd),
Math.max(cdsStart, cdsEnd),
tr.getStrand(),
""); // Create a cds marker
Exon lastExon = null;
int countCodingExons = 0;
for (Exon exon : tr.sortedStrand()) {
if (exon.intersects(cdsEnd)) lastExon = exon;
if (cds.intersects(exon)) countCodingExons++;
}
// Only one coding exon? => No NMD
// Note: I'm assuming that we should have a splice event in a coding part of the transcript for
// NMD to happen.
if (countCodingExons <= 1) return -1;
// Sanity check
if (lastExon == null)
throw new RuntimeException(
"Cannot find last coding exon for transcript '"
+ tr.getId()
+ "' (cdsEnd: "
+ cdsEnd
+ ")\n\t"
+ tr);
// ---
// Find that position of MND_BASES_BEFORE_LAST_JUNCTION before the last exon-exon junction
// ---
int lastExonJunction = tr.isStrandPlus() ? lastExon.getStart() : lastExon.getEnd();
int chrPos[] = tr.baseNumberCds2Pos();
int lastNmdPos = -1;
for (int cdsi = chrPos.length - 1; cdsi >= 0; cdsi--) {
if (chrPos[cdsi] == lastExonJunction) {
if (cdsi > MND_BASES_BEFORE_LAST_JUNCTION)
lastNmdPos = chrPos[cdsi - MND_BASES_BEFORE_LAST_JUNCTION - 1];
else return tr.isStrandPlus() ? 0 : Integer.MAX_VALUE; // Out of CDS range
return lastNmdPos;
}
}
throw new RuntimeException(
"Cannot find last exon junction position for transcript '" + tr.getId() + "'\n\t" + tr);
// return -1;
}
@Test
public void test_01() {
Gpr.debug("Test");
// Test N times
// - Create a random gene transcript, exons
// - Change each base in the exon
// - Calculate effect
for (int i = 0; i < N; i++) {
initSnpEffPredictor();
if (debug)
System.out.println(
"MNP Test iteration: " + i + "\nChromo:\t" + chromoSequence + "\n" + transcript);
else if (verbose)
System.out.println(
"MNP Test iteration: "
+ i
+ "\t"
+ (transcript.isStrandPlus() ? "+" : "-")
+ "\t"
+ transcript.cds());
else Gpr.showMark(i + 1, 1);
if (debug) {
for (Exon exon : transcript.sortedStrand())
System.out.println(
"\tExon: "
+ exon
+ "\tSEQ:\t"
+ (exon.isStrandMinus()
? GprSeq.reverseWc(exon.getSequence())
: exon.getSequence())
.toUpperCase());
}
// For each base in this exon...
for (int pos = 0; pos < chromoSequence.length() - 2; pos++) {
// MNP length
int mnpLen = rand.nextInt(MAX_MNP_LEN) + 2;
int maxMnpLen = chromoSequence.length() - pos;
mnpLen = Math.min(mnpLen, maxMnpLen);
String ref = chromoSequence.substring(pos, pos + mnpLen);
String mnp = createMnp(pos, mnpLen);
analyze(i, pos, ref, mnp);
}
}
System.err.println("");
}
/** Set marker. Add some warnings if the marker relates to incomplete transcripts */
public void setMarker(Marker marker) {
this.marker = marker;
Transcript transcript = getTranscript();
if (transcript != null) {
// Transcript level errors or warnings
addErrorWarningInfo(transcript.sanityCheck(variant));
// Exon level errors or warnings
Exon exon = getExon();
if (exon != null) addErrorWarningInfo(exon.sanityCheck(variant));
}
}
/** Get the simplest string describing the effect (this is mostly used for testcases) */
public String toStringSimple(boolean shortFormat) {
String transcriptId = "";
Transcript tr = getTranscript();
if (tr != null) transcriptId = tr.getId();
String exonId = "";
Exon exon = getExon();
if (exon != null) exonId = exon.getId();
String eff = effect(shortFormat, true, true, false);
if (!eff.isEmpty()) return eff;
if (!exonId.isEmpty()) return exonId;
if (!transcriptId.isEmpty()) return transcriptId;
return "NO EFFECT";
}
public String toString(boolean useSeqOntology, boolean useHgvs) {
// Get data to show
String geneId = "", geneName = "", bioType = "", transcriptId = "", exonId = "", customId = "";
int exonRank = -1;
if (marker != null) {
// Gene Id, name and biotype
Gene gene = getGene();
Transcript tr = getTranscript();
// CDS size info
if (gene != null) {
geneId = gene.getId();
geneName = gene.getGeneName();
bioType = getBiotype();
}
// Update trId
if (tr != null) transcriptId = tr.getId();
// Exon rank information
Exon exon = getExon();
if (exon != null) {
exonId = exon.getId();
exonRank = exon.getRank();
}
// Regulation
if (isRegulation()) bioType = ((Regulation) marker).getCellType();
}
// Add seqChage's ID
if (!variant.getId().isEmpty()) customId += variant.getId();
// Add custom markers
if ((marker != null) && (marker instanceof Custom))
customId += (customId.isEmpty() ? "" : ";") + marker.getId();
// CDS length
int cdsSize = getCdsLength();
String errWarn = error + (error.isEmpty() ? "" : "|") + warning;
String aaChange = "";
if (useHgvs) aaChange = getHgvs();
else aaChange = ((aaRef.length() + aaAlt.length()) > 0 ? aaRef + "/" + aaAlt : "");
return errWarn //
+ "\t"
+ geneId //
+ "\t"
+ geneName //
+ "\t"
+ bioType //
+ "\t"
+ transcriptId //
+ "\t"
+ exonId //
+ "\t"
+ (exonRank >= 0 ? exonRank : "") //
+ "\t"
+ effect(false, false, false, useSeqOntology) //
+ "\t"
+ aaChange //
+ "\t"
+ ((codonsRef.length() + codonsAlt.length()) > 0 ? codonsRef + "/" + codonsAlt : "") //
+ "\t"
+ (codonNum >= 0 ? (codonNum + 1) : "") //
+ "\t"
+ (codonDegeneracy >= 0 ? codonDegeneracy + "" : "") //
+ "\t"
+ (cdsSize >= 0 ? cdsSize : "") //
+ "\t"
+ (codonsAroundOld.length() > 0 ? codonsAroundOld + " / " + codonsAroundNew : "") //
+ "\t"
+ (aasAroundOld.length() > 0 ? aasAroundOld + " / " + aasAroundNew : "") //
+ "\t"
+ customId //
;
}
/**
* Name of the regions hit by a marker
*
* @param marker
* @param showGeneDetails
* @param compareTemplate
* @param id : Only use genes or transcripts matching this ID
* @return
*/
public Set<String> regions(
Marker marker, boolean showGeneDetails, boolean compareTemplate, String id) {
if (Config.get().isErrorOnMissingChromo() && isChromosomeMissing(marker))
throw new RuntimeEOFException("Chromosome missing for marker: " + marker);
boolean hitChromo = false;
HashSet<String> hits = new HashSet<String>();
Markers intersects = query(marker);
if (intersects.size() > 0) {
for (Marker markerInt : intersects) {
if (markerInt instanceof Chromosome) {
hitChromo = true; // OK (we have to hit a chromosome, otherwise it's an error
hits.add(markerInt.getClass().getSimpleName()); // Add marker name to the list
} else if (markerInt instanceof Gene) {
// Analyze Genes
Gene gene = (Gene) markerInt;
regionsAddHit(hits, gene, marker, showGeneDetails, compareTemplate);
// For all transcripts...
for (Transcript tr : gene) {
if ((id == null)
|| gene.getId().equals(id)
|| tr.getId().equals(id)) { // Mathes ID? (...or no ID to match)
// Does it intersect this transcript?
if (tr.intersects(marker)) {
regionsAddHit(hits, tr, marker, showGeneDetails, compareTemplate);
// Does it intersect a UTR?
for (Utr utr : tr.getUtrs())
if (utr.intersects(marker))
regionsAddHit(hits, utr, marker, showGeneDetails, compareTemplate);
// Does it intersect an exon?
for (Exon ex : tr)
if (ex.intersects(marker))
regionsAddHit(hits, ex, marker, showGeneDetails, compareTemplate);
// Does it intersect an intron?
for (Intron intron : tr.introns())
if (intron.intersects(marker))
regionsAddHit(hits, intron, marker, showGeneDetails, compareTemplate);
}
}
}
} else {
// No ID to match?
if (id == null) regionsAddHit(hits, markerInt, marker, showGeneDetails, compareTemplate);
else {
// Is ID from transcript?
Transcript tr = (Transcript) markerInt.findParent(Transcript.class);
if ((tr != null) && (tr.getId().equals(id))) {
regionsAddHit(
hits,
markerInt,
marker,
showGeneDetails,
compareTemplate); // Transcript ID matches => count
} else {
// Is ID from gene?
Gene gene = (Gene) markerInt.findParent(Gene.class);
if ((gene != null) && (gene.getId().equals(id)))
regionsAddHit(
hits,
markerInt,
marker,
showGeneDetails,
compareTemplate); // Gene ID matches => count
}
}
}
}
}
if (!hitChromo) throw new RuntimeException("ERROR: Out of chromosome range. " + marker);
return hits;
}