One of the indispensable pillars to build a foundation Precision Medicine is a reference genome. Over a history of more than 13 years summarized in this article Bộ gen tham chiếu đã được xây dựng như thế nào?,

The process of perfecting the human reference genome has achieved certain achievements. With the continuous progress of science and technology, the reference genome is constantly updated in the direction of becoming more and more accurate, synthesizing more information to represent the most general human genome. But also because of this updating process, many different versions of the reference genome have been born. This may not affect some analyses or may cause large differences in analysis results in the field of bioinformatics. This article will contribute to opening up an overview of the reference genome and the differences between popular versions.

“Shape” of the reference genome

Reference Genome (English: Reference Genome or Reference Assembly) is a digital database of nucleic acid chain, was assembled (assembly) by scientists, considered as a genome of an ideal organism for a species. Because it is assembled from the DNA sequences of a group of experimental individuals, the reference gene does not completely represent the genes of any one individual. Instead, the reference gene provides a mosaic đơn bội of different DNA sequences from each experimental participant.

The simplest format of a reference genome is a file fasta contains sequences of nucleic acids or amino acids, divided into many “contigs” (usually chromosomes). However, this information is only sufficient for one type of analysis: alignment . Other file types – gene annotation files (genome annotation) GTF or GFF allow for more downstream analysis because they show gene properties such as transcriptional region coordinates, exons, introns, etc.

Some versions of the reference genome

Currently, most of the analyses and published articles in the field of human genetics or bioinformatics use two main versions: hg19 (also known as GRCh37 – Genome Reference Consortium Human Build 37, named after the 37th meeting of this conference) or hg38 (GRCh38). However, for each version, there are many additional versions, published by different research units.

hg19 appendices

GRCh37 (NCBI)

GRCh37, The full name is Genome Reference Consortium Human Build 37, named after the 37th meeting of the Genome Reference Consortium conference. The official version was built by Genome Reference Consortium, published on February 27, 2009. The file containing the official reference genome is released and maintained by NCBI. (National Center for Biotechnology Information).

In this appendix, the names of chromosomes 1 to 22 are named NC_00000a.b with a corresponding to 1 to 22 and b is the version number. Similarly, chromosomes X and Y have a=23 and a=24, respectively.

Currently, this appendix has been updated by NCBI to the 13th time, the file name is GCF_000001405.25_GRCh37.p13_genomic.fna.gz with MD5sum 46e212080d30b1a24abec3eab36dbacd.

Official source:

• Updates at hyperlink.

• File containing reference genome FASTA

• Folder containing related files GCF_000001405.25_GRCh37.p13

b37

The Broad Institute created a new reference genome based on NCBI’s GRCh37, named b37. Compared to the original, b37 has a few changes such as consecutive segment names, and low-confidence bases are converted to N characters according to  IUPAC code.

In this appendix, the names of chromosomes 1 to 22 are numbered from 1 to 22, respectively. The X and Y chromosomes are X and Y, respectively.

Official source:

• FTP Broad bundle

humanG1Kv37

This is the appendix used in the analysis of 1000 Genomes Project.

The humanG1Kv37 appendix is ​​equivalent to version b37 but does not contain decoys for human gammaherpesvirus 4 (NC_007605).

In this appendix, chromosomes 1 to 22 are labeled 1 to 22, respectively. Chromosomes X and Y are labeled X and Y, respectively.

Official source:

• File containing reference genome FASTA

• Folder containing related files phase2_reference_assembly_sequence

hg19 (UCSC)

The University of California at Santa Cruz (UCSC) created an hg19 appendix based on GRCh37. The appendix has a reference genome file hg19.fa.gz with MD5sum: 806c02398f5ac5da8ffd6da2d1d5d1a9.

In this appendix, chromosomes 1 to 22 are named chr1 to chr22, respectively. Chromosomes X and Y are chrX and chrY, respectively.

Official source:

• File containing reference genome FASTA

• Folder containing related files Golden Path

hg38 appendices

GRCh38 (NCBI)

GRCh38, The full name is Genome Reference Consortium Human Build 38, named after the 38th meeting of the Genome Reference Consortium conference. The official version was built by Genome Reference Consortium,

published on February 28, 2019. The file containing the official reference genome is released and managed by NCBI (National Center for Biotechnology Information). Currently, this version has been updated by NCBI for the 13th time, the file name is GCA_000001405.28_GRCh38.p13_genomic.fna.gz with MD5sum: f28b7146e0f30efa58447eceb32620a3.

In this appendix, the names of chromosomes 1 to 22 are named CM000a.2 with a corresponding to 663 to 684. Similarly, chromosomes X and Y have a=685 and a=686, respectively.

Official source:

• Updates at this link. File containing the reference genome FASTA

• Folder containing related files GCA_000001405.28_GRCh38.p13

GRCh38 Resource bundle (Broad Institute)

In addition to the reference genome, this appendix also includes standard databases for analyzing human genome sequencing data via GATK – Genome Analysis Toolkit.

The Broad Institute is currently contributing to the global standard for genetic analysis using GRCh38/hg38 through its standard database. The International Genome Sample Resource (IGSR) project also uses this dataset as a basis for analysis in Phase 3.

In this appendix, chromosomes 1 to 22 are labeled chr1 to chr22, respectively. Chromosomes X and Y are chrX and chrY, respectively.

Official source:

• Google Cloud Resource bundle

Detail comparison

In addition to the large change in coordinates, what makes hg38 more useful for analysis than hg19 is the large number of alternate regions. These alternate regions, named “*_alt”, represent sequences that are common in the human population but are quite different from the sequence in hg38. The presence of these alternate regions has made the analysis of different populations around the world more precise.