Skip to content

Hotfix sealed .hic conversion #28

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
AxisAlexNT merged 1 commit into from
Jun 14, 2026
Merged

Hotfix sealed .hic conversion #28

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 29 additions & 8 deletions src/main/java/ru/itmo/ctlab/hict/hict_library/converters/McoolToHictConverter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -184,10 +184,12 @@ private static void applyAssemblyLayoutToOutput(
final long[] componentEndBpExclusive = new long[contigRecords.size()];
final var scaffoldIds = new java.util.LinkedHashMap<String, Long>();
final var sourceChromLayout = readSourceChromLayout(src, selectedResolutions.get(0));
final boolean isJuiceboxAssemblyLayout = layoutPath.getFileName().toString().toLowerCase().endsWith(".assembly");
final boolean allContigNamesResolve = contigRecords.stream()
.allMatch(record -> canResolveSourceChromIndex(sourceChromLayout, record.getContigName()));
final boolean useSingleAssemblyChromosome = !allContigNamesResolve && sourceChromLayout.lengthsBp().length == 1;
if (!allContigNamesResolve && !useSingleAssemblyChromosome) {
final boolean allowUnmappedJuiceboxContigs = !allContigNamesResolve && !useSingleAssemblyChromosome && isJuiceboxAssemblyLayout;
if (!allContigNamesResolve && !useSingleAssemblyChromosome && !allowUnmappedJuiceboxContigs) {
final var missingContigName = contigRecords.stream()
.map(AGPProcessor.ContigAGPRecord::getContigName)
.filter(name -> !canResolveSourceChromIndex(sourceChromLayout, name))
Expand All @@ -199,6 +201,7 @@ private static void applyAssemblyLayoutToOutput(
final Map<String, SourceComponentRange> singleAssemblySourceRanges = useSingleAssemblyChromosome
? readSingleAssemblySourceRanges(layoutPath)
: Map.of();
int unmappedJuiceboxContigs = 0;

for (int i = 0; i < contigRecords.size(); i++) {
final var record = contigRecords.get(i);
Expand All @@ -220,22 +223,35 @@ final var record = contigRecords.get(i);
componentEndBpExclusive[i] = assemblyChromosomeOffsetBp + componentLengthBp;
}
assemblyChromosomeOffsetBp = componentEndBpExclusive[i];
} else if (allowUnmappedJuiceboxContigs && !canResolveSourceChromIndex(sourceChromLayout, record.getContigName())) {
sourceChromIds[i] = -1;
componentStartBp0[i] = 0L;
componentEndBpExclusive[i] = componentLengthBp;
unmappedJuiceboxContigs++;
} else {
sourceChromIds[i] = resolveSourceChromIndex(sourceChromLayout, record.getContigName(), layoutPath);
componentStartBp0[i] = record.getIntraContigStartBpIncl() - 1L;
componentEndBpExclusive[i] = record.getIntraContigEndBpIncl();
}
final long sourceChromLengthBp = sourceChromLayout.lengthsBp()[sourceChromIds[i]];
if (componentStartBp0[i] < 0L || componentEndBpExclusive[i] <= componentStartBp0[i] || componentEndBpExclusive[i] > sourceChromLengthBp) {
throw new IllegalArgumentException(
"Assembly layout component " + record.getContigName() + ":" + record.getIntraContigStartBpIncl() +
"-" + record.getIntraContigEndBpIncl() + " is outside source contig length " + sourceChromLengthBp +
" from " + layoutPath.getFileName()
);
if (sourceChromIds[i] >= 0) {
final long sourceChromLengthBp = sourceChromLayout.lengthsBp()[sourceChromIds[i]];
if (componentStartBp0[i] < 0L || componentEndBpExclusive[i] <= componentStartBp0[i] || componentEndBpExclusive[i] > sourceChromLengthBp) {
throw new IllegalArgumentException(
"Assembly layout component " + record.getContigName() + ":" + record.getIntraContigStartBpIncl() +
"-" + record.getIntraContigEndBpIncl() + " is outside source contig length " + sourceChromLengthBp +
" from " + layoutPath.getFileName()
);
}
}
contigLengthBp[i] = componentLengthBp;
contigScaffoldIds[i] = scaffoldIds.computeIfAbsent(record.getScaffoldName(), ignored -> (long) scaffoldIds.size());
}
if (unmappedJuiceboxContigs > 0) {
logConsumer.accept(
"Assembly layout contains " + unmappedJuiceboxContigs +
" contig(s) that are absent from source .mcool chromosome metadata; keeping them as hidden zero-bin contigs."
);
}

final long totalAssemblyLengthBp = Arrays.stream(contigLengthBp).sum();
if (totalAssemblyLengthBp <= 0L) {
Expand All @@ -259,6 +275,11 @@ final var record = contigRecords.get(i);
final long[] sourceBinStarts = src.int64().readArray("/resolutions/" + resolution + "/bins/start");
final long[] sourceBinEnds = src.int64().readArray("/resolutions/" + resolution + "/bins/end");
for (int i = 0; i < contigRecords.size(); i++) {
if (sourceChromIds[i] < 0) {
startBins[i] = 0L;
lengthBins[i] = 0L;
continue;
}
final var range = resolveSourceBinRange(
resolution,
sourceChromIds[i],
Expand Down
47 changes: 47 additions & 0 deletions src/test/java/ru/itmo/ctlab/hict/hict_library/converters/McoolToHictConverterTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -155,6 +155,53 @@ void reorderedJuiceboxAssemblyUsesHeaderOrderForSingleHictkSourceOffsets() throw
}
}

@Test
void juiceboxAssemblyKeepsContigsMissingFromMultiChromSourceHidden() throws Exception {
final var mcool = tempDir.resolve("partial-source.mcool");
final var assembly = tempDir.resolve("partial-layout.assembly");
final var output = tempDir.resolve("partial-output.hict.hdf5");

writeSyntheticMcool(mcool);
Files.writeString(
assembly,
String.join(
System.lineSeparator(),
">ctgA 1 2500",
">ctgB 2 3600",
">ctgC 3 1500",
"1",
"2",
"3"
) + System.lineSeparator()
);

new McoolToHictConverter().convert(
new ConversionOptions(
mcool,
output,
List.of(1_000L),
64,
0,
ConversionOptions.CompressionAlgorithm.DEFLATE,
assembly.toString(),
false,
1
),
ignored -> {
}
);

try (final var reader = HDF5Factory.openForReading(output.toFile())) {
assertArrayEquals(new String[]{"ctgA", "ctgB", "ctgC"}, reader.string().readArray(getContigNameDatasetPath()));
assertArrayEquals(new long[]{3L, 4L, 0L}, reader.int64().readArray(getContigLengthBinsDatasetPath(1_000L)));

final long[][] basisAtu = reader.int64().readMatrix(getBasisATUDatasetPath(1_000L));
assertEquals(2, basisAtu.length);
assertArrayEquals(new long[]{0L, 0L, 3L, 1L}, basisAtu[0]);
assertArrayEquals(new long[]{0L, 3L, 7L, 1L}, basisAtu[1]);
}
}

private static void writeSyntheticMcool(final Path path) {
HDF5LibraryInitializer.initializeHDF5Library();
try (final var writer = HDF5Factory.open(path.toFile())) {
Expand Down
10 changes: 5 additions & 5 deletions ...itmo/ctlab/hict/hict_server/handlers/conversion/HicAssemblyConversionIntegrationTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,9 +16,9 @@
import static org.junit.jupiter.api.Assertions.assertTrue;

class HicAssemblyConversionIntegrationTest {
private static final Path HIC = Path.of("/mnt/Models/HiCT/data/DNAZoo/AedisAegypti/AGWG.draft.hic");
private static final Path ASSEMBLY = Path.of("/mnt/Models/HiCT/data/DNAZoo/AedisAegypti/AGWG.draft.assembly");
private static final Path FASTA = Path.of("/mnt/Models/HiCT/data/DNAZoo/AedisAegypti/AGWG.draft.fasta.gz");
private static final Path HIC = Path.of("/mnt/Models/HiCT/data/DNAZoo/AedesAegypti/AaegL5.0.hic");
private static final Path ASSEMBLY = Path.of("/mnt/Models/HiCT/data/DNAZoo/AedesAegypti/AaegL5.0.assembly");
private static final Path FASTA = Path.of("/mnt/Models/HiCT/data/DNAZoo/AedesAegypti/AaegL5.0.fasta.gz");

@TempDir
Path tempDir;
Expand All @@ -44,7 +44,7 @@ void convertsAndOpensHicWithJuiceboxAssemblyAndFasta() throws Exception {
new ConversionOptions(
HIC,
output,
List.of(),
List.of(2_500_000L),
8_192,
6,
ConversionOptions.CompressionAlgorithm.DEFLATE,
Expand Down Expand Up @@ -80,7 +80,7 @@ void convertsAndOpensHicWithJuiceboxAssemblyAndFasta() throws Exception {
chunkedFile.getAssemblyInfo().contigs().size(),
chunkedFile.getContigTree().getOrderedContigList().size()
);
assertTrue(chunkedFile.getContigTree().getOrderedContigList().size() > 1);
assertTrue(chunkedFile.getContigTree().getOrderedContigList().size() > 3);
}

private static Path findLocalHictkBinary() throws IOException {
Expand Down
2 changes: 1 addition & 1 deletion version.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1 +1 @@
1.0.175-5ddcf3c-webui_a609177
1.0.176-533a1f0-webui_a609177
Loading