Name (link) Description 特点
    1000 Genomes reference Used in this course. The 1000g reference names chromosomes as follows (chr1, chr2, .., chr22, chrX, chrY, chrM). This reference includes “decoy” sequences (mostly low complexity sequences) that have been added to the standard genome build sequence. This reduces misalignment of reads that would otherwise get placed somewhere they don’t belong. The developer of the BWA aligner documents use of this version of the reference genome. This reference includes the alternative contigs. BWA文档中使用
    Ensembl reference Ensembl names the chromosomes as follows (1, 2, .., 22, X, Y, MT). The names of some unplaced contigs also differ. This reference does NOT have the decoy sequences. This reference includes the alternative contigs. 染色体号没有“chr”前缀
    UCSC reference The UCSC reference names chromosomes as follows (chr1, chr2, .., chr22, chrX, chrY, chrM). This reference does NOT have the decoy sequences. This reference includes the alternative contigs.
    NCBI reference NCBI names the chromosomes as follows (chr1, chr2, .., chr22, chrX, chrY, chrMT). This reference does NOT include the decoy sequences. This reference includes the alternative contigs. The major annotation centers such as UCSC and Ensembl start with raw files from NCBI (Various Human Assemblies). Most other people do not use these NCBI files directly but rather get a version of the files from UCSC, Ensembl, etc.
    Genomic Data Commons (GDC) reference The GDC reference names chromosomes as follows (chr1, chr2, .., chr22, chrX, chrY, chrM). The GDC created their own version of the reference for harmonized analysis of the TCGA and other large cancer sequencing projects. This reference includes “decoy” sequences. This reference does NOT include the alternative contigs. Unique to this reference is the inclusion of several virus sequences for viruses with known or suspected roles in cancer (e.g. HPV, EBV, etc.). 没有contig， 包含病毒序列

    from: <pmbio: Reference Genome>