Impute Benchmark

Allelica supports users with a simple process to analyze chromosomal variance. This support is provided within Illumina's BaseSpace.

The Science Behind Allelica's Low Pass Impute App

These documents provide further information on the computational approach provided by Allelica's Low Pass Impute app. The following aspects are covered:

How Accurate is the Imputation?

Details of the analysis

Three individual High Coverage sequenced genomes (30X of coverage depth) of different ethnicity from 1000 genomes database were selected:

NA12878: European Participant

NA19700: African Participant

HG00479: Chinese Participant

Through the samtools program, each of the three genomes has been downsampled to a range of different coverage depths, ranging from 0.2X to 8X. This step generated for each individual a set of 9 downsampled genomes of different coverage depth.
The three sets of downsampled genomes have been imputed using the program GLIMPSE. A set of 2501 haplotypes from 1000 Genomes has been used as reference panel in the imputation process. The individuals under analysis (i.e., NA12878, NA19700, and HG00479) have been removed from the reference panel before imputation.
In order to mimic quality controls, nucleotide variants displaying a maximum genotype probability lower than 0.9 were removed from all imputed genomes.
Imputation accuracy has been assessed for each individual and each downsampled coverage depth. Since imputation accuracy is strongly dependent on the allele frequency, this has been quantified for a range of different alternate allele frequency bins. Imputation accuracy has been measured by quantifying for each frequency bin the concordance (a measure of the percentage of SNPs that are measured as identical) between each downsample genome and the original genome. Figures report the dependence of Concordance from alternate allele frequency bins for 4 illustrative coverage depths (0.2X, 0.5X, 1X, and 4X), while Tables report concordance values for all coverage depths analyzed.

Results for Different Minor Allele Frequency

European

Alternate

Allele

Frequency

Coverage 0.2X

Coverage 0.4X

Coverage 0.5X

Coverage 0.6X

Coverage 0.8X

Coverage 1.0X

Coverage 2.0X

Coverage 4.0X

0.002

0.153

0.323

0.363

0.397

0.439

0.477

0.571

0.666

0.003

0.308

0.497

0.551

0.589

0.647

0.663

0.73

0.782

0.004

0.419

0.635

0.672

0.7

0.738

0.758

0.801

0.841

0.008

0.583

0.741

0.782

0.793

0.821

0.837

0.864

0.878

0.015

0.769

0.856

0.872

0.881

0.892

0.9

0.914

0.925

0.033

0.881

0.928

0.935

0.94

0.946

0.95

0.955

0.958

0.073

0.946

0.967

0.97

0.972

0.973

0.975

0.976

0.977

0.146

0.97

0.98

0.982

0.983

0.984

0.248

0.979

0.985

0.986

0.987

0.348

0.982

0.987

0.988

0.449

0.985

0.988

0.989

0.988

0.549

0.987

0.99

0.991

0.649

0.988

0.992

0.993

0.992

0.993

0.992

Results for Different Ancestries

African

Alternate

Allele

Frequency

Coverage 0.2X

Coverage 0.4X

Coverage 0.5X

Coverage 0.6X

Coverage 0.8X

Coverage 1.0X

Coverage 2.0X

Coverage 4.0X

Coverage 8.0X

0.002

0.212

0.385

0.432

0.475

0.526

0.559

0.640

0.720

0.797

0.003

0.344

0.601

0.655

0.697

0.760

0.779

0.831

0.863

0.884

0.004

0.488

0.745

0.780

0.811

0.850

0.860

0.892

0.907

0.92

0.008

0.665

0.839

0.865

0.883

0.902

0.912

0.929

0.938

0.943

0.015

0.789

0.904

0.920

0.930

0.942

0.947

0.956

0.962

0.965

0.033

0.861

0.940

0.950

0.956

0.962

0.967

0.973

0.977

0.978

0.073

0.913

0.960

0.966

0.970

0.974

0.977

0.982

0.983

0.984

0.146

0.945

0.973

0.976

0.979

0.981

0.983

0.986

0.987

0.248

0.962

0.979

0.982

0.983

0.984

0.985

0.987

0.348

0.972

0.982

0.984

0.985

0.986

0.987

0.988

0.987

0.449

0.976

0.984

0.985

0.987

0.986

0.549

0.979

0.986

0.987

0.988

0.989

0.990

0.989

0.649

0.984

0.989

0.990

0.991

0.992

Alternate

Allele

Frequency

Coverage 0.2X

Coverage 0.4X

Coverage 0.5X

Coverage 0.6X

Coverage 0.8X

Coverage 1.0X

Coverage 2.0X

Coverage 4.0X

Coverage 8.0X

0.002

0.005

0.037

0.064

0.086

0.124

0.163

0.320

0.480

0.662

0.003

0.070

0.173

0.209

0.248

0.315

0.347

0.514

0.656

0.766

0.004

0.135

0.259

0.319

0.372

0.451

0.506

0.652

0.743

0.818

0.008

0.142

0.321

0.398

0.455

0.535

0.595

0.734

0.805

0.855

0.015

0.396

0.594

0.656

0.691

0.742

0.776

0.847

0.889

0.915

0.033

0.687

0.825

0.856

0.869

0.892

0.907

0.934

0.948

0.956

0.073

0.852

0.926

0.938

0.945

0.956

0.962

0.969

0.973

0.975

0.146

0.917

0.960

0.966

0.970

0.975

0.978

0.981

0.983

0.984

0.248

0.944

0.971

0.974

0.977

0.979

0.982

0.985

0.986

0.348

0.960

0.978

0.979

0.981

0.983

0.985

0.986

0.987

0.449

0.967

0.981

0.983

0.984

0.985

0.986

0.988

0.549

0.975

0.984

0.985

0.986

0.987

0.988

0.990

0.991

0.649

0.978

0.986

0.987

0.989

0.990

0.991

PreviousImpute

Last updated 3 years ago