Package

io.projectglow

bgen

Permalink

package bgen

Visibility
  1. Public
  2. All

Type Members

  1. class BgenFileFormat extends FileFormat with DataSourceRegister with Serializable with GlowLogging

    Permalink
  2. class BgenRecordWriter extends GlowLogging

    Permalink

    Writes BGEN records.

  3. class BgenRowToInternalRowConverter extends GlowLogging

    Permalink

    Converts BgenRows into InternalRow with a given required schema.

    Converts BgenRows into InternalRow with a given required schema. During construction, this class will throw an IllegalArgumentException if any of the fields in the required schema cannot be derived from a BGEN record.

  4. class BigBgenDatasource extends BigFileDatasource with DataSourceRegister

    Permalink
  5. case class BigBgenOptions(bitsPerProb: Int, maxPloidy: Int, defaultPloidy: Int, defaultPhasing: Boolean) extends Product with Serializable

    Permalink
  6. class ComDatabricksBgenFileFormat extends BgenFileFormat with ComDatabricksDataSource

    Permalink
  7. class ComDatabricksBigBgenDatasource extends BigBgenDatasource with ComDatabricksDataSource

    Permalink
  8. case class FractionalPartIndex(fractionalPart: Double, index: Int) extends Product with Serializable

    Permalink
  9. case class GenotypeCharacteristics(numAlleles: Int, phased: Boolean, ploidy: Int) extends Product with Serializable

    Permalink
  10. class InternalRowToBgenRowConverter extends GlowLogging

    Permalink

    Converts internal rows to BGEN rows.

    Converts internal rows to BGEN rows. Includes logic to infer phasing and ploidy if missing (eg. when converting from VCF rows with no GT field), using the number of alleles and the number of posterior probabilities.

    - If phasing and ploidy are missing, we assume ploidy is defaultPloidy. - If phasing is missing:

    • If no posterior probabilities are present, we assume phasing is defaultPhasing.
    • If the number of posterior probabilities matches the case that the probability represents:
      • Either phased or unphased data: we assume phasing is defaultPhasing.
      • Phased data: we assume the data is phased.
      • Unphased data: we assume the data is unphased.
      • Neither: we throw an exception - If ploidy is missing:
    • If no posterior probabilities are present, we assume ploidy is defaultPloidy.
    • If phased, we try to calculate the ploidy directly.
    • If unphased, we try to find the ploidy between [1, maxPloidy].
    Exceptions thrown

    IllegalStateException if phasing or ploidy cannot be inferred or a single row contains both unphased and phased data.

  11. case class ProbabilityBlockSize(probabilitiesPerBlock: Int, numBlocks: Int) extends Product with Serializable

    Permalink
  12. sealed trait SNPBlockCompression extends AnyRef

    Permalink
  13. class SortByDescendingFractionalPart extends Comparator[FractionalPartIndex]

    Permalink

Value Members

  1. object BgenFileFormat extends HlsEventRecorder with Serializable

    Permalink
  2. object BgenRecordWriter

    Permalink
  3. object BgenSchemaInferrer

    Permalink

    Infers the schema of a set of BGEN files from the user-provided options and the header of each file.

    Infers the schema of a set of BGEN files from the user-provided options and the header of each file.

    Currently the implementation is very simple. It checks if sample IDs are defined in either a .sample file or in any of the headers. If so, it returns a fixed schema that includes a sampleId genotype field. If not, it returns the same schema without a sampleId field.

  4. object BigBgenDatasource extends HlsEventRecorder

    Permalink
  5. object Zlib extends SNPBlockCompression with Product with Serializable

    Permalink
  6. object Zstd extends SNPBlockCompression with Product with Serializable

    Permalink

Ungrouped