/**
* This function figures out the consequence of a DNA duplication that does not begin at position
* zero of a codon, i.e., that is not right in the frame. This can result in a delins or in a
* simple dup depending on the surrounding sequence. For example, imagine we have the following
* duplication. ggaggaggaggaggaggagga (add another gga). but the frame is gag-gag-gag-...with
* GAG=Glu/E. Then the effect of the duplication is to add another E to the aminoacid sequence.
* The variable frame_s is 2 in this case, because gga starts at nucleotide 2 (zero-based) of the
* GAG codon.
*
* @param trmdl The affected transcript
* @param exonNumber One-based number of the exon
* @param cDNAannot Annotation of the cDNA, e.g., c.769_771dupTTC
* @param var the duplicated sequence
* @param endpos the end position of the duplication (zero-based)
* @param frame_s the location within the frame (0,1,2) in which mutation occurs
*/
private static String shiftedInFrameDuplication(
TranscriptModel trmdl,
int exonNumber,
String cDNAanno,
String var,
int endpos,
int aaVarStartPos,
int frame_s)
throws AnnotationException {
Translator translator = Translator.getTranslator(); /* Singleton */
int len = var.length();
if ((len % 3) != 0) {
String s =
String.format(
"[ERROR] DuplicationAnnotation:shiftedInFrameDuplication - "
+ "variant length not a multiple of 3: %s (len=%d)",
var, len);
throw new AnnotationException(s);
}
int aalen = len / 3;
String dna = trmdl.getCdnaSequence();
int start = endpos - var.length();
String prefix = dna.substring(start - frame_s, start);
// System.out.println("prefix = " + prefix + ", frame_s="+frame_s + ", var="+var + " endpos=" +
// endpos);
// System.out.println(cDNAanno);
String rest = dna.substring(start, start + len);
String wt = prefix + rest;
String mut = prefix + var + rest;
String wtaa = translator.translateDNA(wt);
String mutaa = translator.translateDNA(mut);
String annot;
if (mutaa.startsWith(wtaa) && (mutaa.indexOf(wtaa, aalen)) > 0) {
annot =
String.format(
"%s:exon%d:%s:p.%s%ddup", trmdl.getName(), exonNumber, cDNAanno, wtaa, aaVarStartPos);
} else {
annot =
String.format(
"%s:exon%d:%s:p.%s%ddelins%s",
trmdl.getName(), exonNumber, cDNAanno, wtaa, aaVarStartPos, mutaa);
}
return annot;
}
/**
* Annotates an insertion variant that is an duplication. The methods of this class are called
* from {@link jannovar.annotation.InsertionAnnotation InsertionAnnotation} if that class
* determines that the insertion is equal to the preceding nucleotides in the reference sequence.
* That is, in addition to the conditions for a insertion variant, the duplication variant
* requires a similar sequence to the insertion (before or) after the insertion.
*
* <p>There are two possible duplication insertions with or without frameshift causation. e.g.
* inserting an additional 'C' in the sequence 'ACC,GAG' at position 2 would cause a frameshift,
* whereas insertion of 'CCG' at position 2 just inserts an additional triple 'ACC GCC GAG'.
*
* <p>if (var.length() % 3 == 0) { /* ORF CONSERVING if(startPosMutationInCDS.length() % 3 == 0){
* /* SIMPLE DUPLICATION OF CODONS } else { /* substitution from original AA to AAs
* if(wtaa.equals("*")) { /* Mutation affects the wildtype stop codon int idx =
* varaa.indexOf("*"); if (idx < 0) { } /* Substitution } }else { /* FRAMESHIFT * short
* p.(Arg97fs)) denotes a frame shifting change with Arginine-97 as the first affected amino acid
* }
*
* @param trmdl The transcriptmodel / gene in which the current mutation is contained
* @param frame_s the location within the frame (0,1,2) in which mutation occurs
* @param wtnt3 The three nucleotides of codon affected by start of mutation
* @param var alternate nucleotide sequence (the duplication)
* @param startpos The startposition of the duplication (zero based)
* @param endpos The endposition of the duplication (zero based)
* @param exonNumber Number (one-based) of affected exon.
* @return an {@link jannovar.annotation.Annotation Annotation} object representing the current
* variant
* @throws AnnotationException
*/
public static Annotation getAnnotation(
TranscriptModel trmdl,
int frame_s,
String wtnt3,
String var,
int startpos,
int endpos,
int exonNumber)
throws AnnotationException {
String annot;
Annotation ann;
Translator translator = Translator.getTranslator(); /* Singleton */
int refcdsstart = trmdl.getRefCDSStart();
/**/
int newpos = shiftToThreePrime(trmdl, var, startpos, endpos);
if (newpos != startpos) {
startpos = newpos;
endpos = startpos + var.length() - 1;
}
int cdsEndPos = endpos - refcdsstart + 1;
int cdsStartPos = cdsEndPos - var.length() + 1;
/**
* aavarpos is now the FIRST position (one-based) of the amino-acid sequence that was
* duplicated.
*/
int aaVarStartPos =
cdsStartPos % 3 == 0
? (int) Math.floor(cdsStartPos / 3)
: (int) Math.floor(cdsStartPos / 3) + 1;
// debugDuplication(trmdl,frame_s, wtnt3, var, startpos, exonNumber,aaVarStartPos);
/* get coding DNA HGVS string */
String canno;
if (var.length() == 1) canno = String.format("c.%ddup%s", cdsStartPos, var);
else canno = String.format("c.%d_%ddup%s", cdsStartPos, cdsEndPos, var);
/* now create the protein HGVS string */
/* generate in-frame snippet for translation and correct for '-'-strand */
if (trmdl.isMinusStrand()) {
/* Re-adjust the wildtype nucleotides for minus strand */
wtnt3 = trmdl.getWTCodonNucleotides(startpos - 1 + ((3 - (var.length() % 3)) % 3), frame_s);
}
String varnt3 = getVarNt3(trmdl, wtnt3, var, frame_s);
String wtaa = translator.translateDNA(wtnt3);
String varaa = translator.translateDNA(varnt3);
if (var.length() % 3 == 0) {
/* ORF CONSERVING */
if ((cdsStartPos - 1) % 3 == 0) {
/* SIMPLE DUPLICATION OF CODONS, e.g., nucleotide position 4 starts a
codon, and (4-1)%3==0.*/
String wtaaDupStart = translator.translateDNA(var.substring(0, 3));
String wtaaDupEnd = translator.translateDNA(var.substring(var.length() - 3));
if (var.length() == 3) {
// Three nucleotides affected, inframe, single aminoacid duplication.
annot =
singleAminoAcidInframeDuplication(
trmdl.getName(), exonNumber, canno, wtaaDupStart, aaVarStartPos);
} else {
int aaEndPos =
aaVarStartPos + (var.length() / 3) - 1; /* last amino acid of duplicated WT seq. */
annot =
multipleAminoAcidInframeDuplication(
trmdl.getName(),
exonNumber,
canno,
wtaaDupStart,
aaVarStartPos,
wtaaDupEnd,
aaEndPos);
}
ann = new Annotation(trmdl, annot, VariantType.NON_FS_DUPLICATION, cdsStartPos);
} else {
/* substitution from original AA to AAs */
if (wtaa.equals("*")) {
/* Mutation affects the wildtype stop codon */
int idx = varaa.indexOf("*");
if (idx < 0) {
annot =
String.format(
"%s:exon%d:%s:p.*%d%sext*?",
trmdl.getName(), exonNumber, canno, aaVarStartPos, varaa);
} else {
annot =
String.format(
"%s:exon%d:%s:p.*%ddelins%s",
trmdl.getName(), exonNumber, canno, aaVarStartPos, varaa.substring(0, idx + 1));
}
} else {
/* substitution starts not on frame */
annot =
shiftedInFrameDuplication(
trmdl, exonNumber, canno, var, endpos, aaVarStartPos, frame_s);
}
/* Substitution */
ann = new Annotation(trmdl, annot, VariantType.NON_FS_DUPLICATION, cdsStartPos);
}
} else {
/* FRAMESHIFT
* short p.(Arg97fs)) denotes a frame shifting change with Arginine-97 as the first affected amino acid */
annot =
String.format(
"%s:exon%d:%s:p.%s%dfs", trmdl.getName(), exonNumber, canno, wtaa, aaVarStartPos);
// System.out.println("FS wtaa="+wtaa);
// System.out.println(annot);
ann = new Annotation(trmdl, annot, VariantType.FS_DUPLICATION, cdsStartPos);
}
return ann;
}