1000 Genomes 2504 phase 3 panel samples sequenced to high coverage ================================================================== This policy refers to 30x Illumina NovaSeq sequencing of 2504 samples from the 1000 Genomes project phase 3 sample set. These data were generated at the New York Genome Center with funds provided by NHGRI Grant 3UM1HG008901-03S1. Please email service@nygenome.org with questions or interest in undertaking collaborative analysis of this dataset. All cell lines were obtained from the Coriell Institute for Medical Research and were consented for full public release of genomic data. Please see Coriell (https://www.coriell.org) for more information about specific cell lines. The following cell lines/DNA samples were obtained from the NIGMS Human Genetic Cell Repository at the Coriell Institute for Medical Research: [NA06984, NA06985, NA06986, NA06989, NA06994, NA07000, NA07037, NA07048, NA07051, NA07056, NA07347, NA07357, NA10847, NA10851, NA11829, NA11830, NA11831, NA11832, NA11840, NA11843, NA11881, NA11892, NA11893, NA11894, NA11918, NA11919, NA11920, NA11930. NA11931, NA11932, NA11933, NA11992, NA11994, NA11995, NA12003, NA12004, NA12005, NA12006, NA12043, NA12044, NA12045, NA12046, NA12058, NA12144, NA12154, NA12155, NA12156, NA12234, NA12249, NA12272, NA12273, NA12275, NA12282, NA12283, NA12286, NA12287, NA12340, NA12341, NA12342, NA12347, NA12348, NA12383, NA12399, NA12400, NA12413,, NA12414, NA12489, NA12546, NA12716, NA12717, NA12718, NA12748, NA12749, NA12750, NA12751, NA12760, NA12761, NA12762, NA12763, NA12775, NA12776, NA12777, NA12778, NA12812, NA12813, NA12814, NA12815, NA12827, NA12828, NA12829, NA12830, NA12842, NA12843, NA12872, NA12873, NA12874, NA12878, NA12889, NA12890]. If using this data please acknowledge that: “These data were generated at the New York Genome Center with funds provided by NHGRI Grant 3UM1HG008901-03S1.” Additionally, if data from any of the cell lines listed were used, then they should be cited specifically as coming from NIGMS. For more general information about the consent for the samples in this data set, please see http://www.internationalgenome.org/about#g1k_data_reuse. For general enquiries, please contact info@1000genomes.org. TOPMed Freeze 9 Variant Callset ================================= Overview ------------- Data freeze 9 is based on variant discovery using 143,415 participants in 37 TOPMed studies, 60,540 samples sequenced by NHGRI Centers for Common Disease Genomics (CCDG) and 2,504 1000 Genomes Project samples deeply sequenced by the New York Genome Center. For this data freeze we have relaxed the sequencing quality control criteria to include samples with up to 10% estimated DNA sample contamination (previously 3%) and at least 15 x average coverage (previously 30 x). There are no thresholds on the fractions of the genome covered to at least 10 x or 20 x. In addition to 139,088 samples sequenced in the TOPMed program (after excluding samples with no consent), this freeze includes 19,382 samples from TOPMed studies that were sequenced by CCDG for a total of 158,470 TOPMed study samples, plus 2,504 1000 Genomes Project samples and a grand total of 160,974 samples. The genotypes for samples from CCDG-only studies will be returned separately to CCDG. Variant Calling, Filtering, and Phasing ------------------------------------- The latest version of the GotCloud/vt pipeline is used for calling both SNPs and indels, focusing on bi-allelic SNPs and indels. Briefly, candidate SNP and indel sites are called from the aligned reads for each sample. These are combined and normalized across all samples, accounting for tandem repeats. Genotypes for each individual are called on the union of all variant sites and merged across samples. Variant filtering uses a support vector machine (SVM) classifier. HapMap3 and Omni2.5 polymorphic SNPs serve as positive training labels and variants showing 3 or more Mendelian inconsistencies among 38,866 samples that are duplicates or members of nuclear families serve as negative labels. In addition to improved SVM filtering using local LD information, variants in the NCBI GWAS catalog are flagged in the INFO field. GWAS variants which pass the SVM filter are marked as PASS. Haplotypes were statistically phased using Eagle 2.4 (Dec 13, 2017) in approximately 1 Mbp chunks and ligated using m3vcftools. FILTER reasons --------------------- The BCF files are VCF 4.2 compatible, thus have self-explanatory descriptions of the FILTER, INFO, and FORMAT fields. All files summarize site filtering information in the FILTER column. For SNPs and Indels, possible FILTER values include: * PASS - The variant site passes all filters * SVM - The variant shows an SVM score more negative than -0.5 (hence fails the SVM filter) * CEN - The variant falls in a centromeric region with inferred reference sequence * DUP2 - The variant shows more than 2 percent discordance between duplicates * TRI2 - The variant shows more than 2 percent Mendelian inconsistencies * MIS2 - The variant shows more than 2 percent of samples below depth 10 * DISC - The variant shows more than 5 percent duplicate or Mendelian inconsistencies * EXHET - The variant has excess heterozygosity and HWE p-value less than 1e-6 * CHRXHET - The variant has excess heterozygosity in male chrX (meaning that the male diploid model likelihood is much higher than male haploid model). This filter is applied only in the non-pseudo-autosomal region of chromosome X. INFO fields --------------- The INFO column in VCF contains the following information. The VCF header contains self-explanatory descriptions of these fields as well. Primary INFO fields include: * AVGDP - Average depth per sample without read-level filtering * AC - Alternate allele counts * AN - Total number of alleles * GWAS - Variant in the GWAS catalog; INFO field shows the previous FILTER value * SVM - Score from the Milk-SVM classifier for variant filtering (failure threshold is -0.5) * ANN - (Annotated sites only) Variant functional annotations from snpEff (details below) Additional INFO fields (used for SVM filtering) include: * FIBC_P - Pooled excess homozygosity estimate (inbreeding coefficient): "+" means excess homozygosity, "-" means excess heterozygosity compared to HWE * HWE_SLP_P - Signed log 10 p-value for testing Hardy-Weinberg equilbrium ignoring population substructure (matches sign of FIBC_P) * FIBC_I - Excess homozygosity (inbreeding coefficient) adjusted for population structure and calculated from genotype likelihoods: "+" means excess homozygosity, "-" means excess heterozygosity compared to HWE * HWE_SLP_I - Signed log 10 p-value for testing Hardy-Weinberg equilbrium after adjusting for population substructure (matches sign of FIBC_I) * MAX_IF - Maximum individual-specific alternate allele frequency * MIN_IF - Minimum individual-specific alternate allele frequency * LLK0 - Null likelihood (for debugging purposes) * BETA_IF - Coefficients for intercept and each eigenvector used to calculate individual specific allele frequency * ABE - Mean of allele balance criterion across all heterozygous genotype calls (higher than 0.5 indicates bias towards the reference allele) * ABZ - Mean of Z-scores for allele balance across all heterozygous genotype calls (positive indicates bias towards the reference allele) * BQZ - Z-score for testing association between base qualities and alleles * MQZ - Z-score for testing association between mapping quality and alleles * CYZ - Z-score for testing association between position in the sequence read and alleles * STZ - Z-score for testing association between strand and alleles * NMZ - Z-score for testing association between number of mismatches per read and alleles * IOR - Estimated inflation of base quality in log 10 scale * NM0 - Average number of mismatches in reads showing the reference allele * NM1 - Average number of mismatches in reads showing the non-reference allele * FMIS10 - Fraction of samples with depth below 10 x * MAX_HM3_R2 - Largest r2 to any HapMap3 variant within 1Mb * LOG_HM3_BUDDY_03 - Number of HapMap3 variants within 1Mb with r2 >= 0.3, in log10(1+x) scale * LOG_HM3_BUDDY_05 - Number of HapMap3 variants within 1Mb with r2 >= 0.5, in log10(1+x) scale * LOG_HM3_BUDDY_08 - Number of HapMap3 variants within 1Mb with r2 >= 0.8, in log10(1+x) scale * LOG_HM3_DIST_03 - Distance to nearest HapMap3 variant within 1Mb with r2 >= 0.3, log10(1+x) scale * LOG_HM3_DIST_05 - Distance to nearest HapMap3 variant within 1Mb with r2 >= 0.5, log10(1+x) scale * LOG_HM3_DIST_08 - Distance to nearest HapMap3 variant within 1Mb with r2 >= 0.8, log10(1+x) scale * DUP_NH_ALL - Number of duplicate samples with genotype depth >= 10 and one or more non-reference alleles. This is used as the denominator to calculate duplicate discordance. * DUP_NH_DIS - Number of duplicate samples with genotype depth >= 10 and a discordant genotype. This is used as the numerator to calculate duplicate discordance. * TRI_NH_ALL - Number of trios (1) whose genotype depths are all 10 or greater, (2) genotypes are informative for evaluating Mendelian consistency, and (3) the genotypes are not all homozygous reference. This is used as the denominator to calculate Mendelian discordance. * TRI_NH_DIS - Number of trios (1) whose genotype depths are all 10 or greater, and (2) genotypes indicate Mendelian inconsistency. This is used as the numerator to calculate Mendelian discordance. Formatting details for annotations: Variants are annotated with snpEff 4.3t (build 2017-11-24 10:18), using the GRCh38.86 database. * ANN - Variant functional annotations from snpEff: 'Allele | Annotation | Annotation_Impact | Gene_Name | Gene_ID | Feature_Type | Feature_ID | Transcript_BioType | Rank | HGVS.c | HGVS.p | cDNA.pos / cDNA.length | CDS.pos / CDS.length | AA.pos / AA.length | Distance | ERRORS / WARNINGS / INFO'