MoMI-G tools

The MoMI-G package includes a set of scripts (MoMI-G tools) that converts a VCF file into the variation graph format.

If you want to run MoMI-G with your own dataset, use our custom scripts scripts/vcf2xg.sh to generate pcf and xg dataset. It requires VG, ruby, bash, and samtools.

Software	Dataset	Supported SV type
Sniffles	for PacBio/Nanopore	INV/DEL/DUP/TRA/(INS)
10X LongRanger	for 10X	INV/DEL/DUP/TRA
SURVIVOR	for merging SV calls	INV/DEL/DUP/TRA

$ bash vcf2xg.sh test.vcf test_output /bin/vg hg[19|38]

Otherwise, you can specify your own reference file as follows.

$ bash vcf2xg.sh test.vcf test_output /bin/vg /data/hs37d5.fa

You can use singularity instead.

$ singularity -s run docker://momigteam/momig-tools:master test.vcf test_output vg hg38

After that, these files are required to be mounted on static/ folder. Also, you should modify config.yaml and Dockerfile.backend in MoMI-G directory.

• static/

  • config.yaml: a configuration file
  • *.xg: an index of a variation graph, generated by [vg](https://github.com/vgteam/vg)
  • *.pcf: a pair of coordinate format file: required to display variants on Feature Table or Circos Plot.
  • *.gam(optional): read alignments on the graph
  • *.gam.index(optional): index of gam

This is an example of config.yaml and Dockerfile. \* is just an example and you need to replace it to the actual file name.

bin:
  vg: "vg"
  vg_tmp: "vg"
  vg_volume_prefix: ""
  graphviz: "dot"
  fa22bit: "faToTwoBit"
  bigbed: "bedToBigBed"
reference:
  chroms: "static/GRCh.json"
  data:
    - name: "hg19"
      features:
        - name: 'gene_annotation'
          url: "static/gencode.v27lift37.basic.annotation.gff3"
          chr_prefix: "chr"
    - name: "hg38"
      features:
        - name: 'gene_annotation'
          url: "static/gencode.v27.basic.annotation.gff3"
          chr_prefix: "chr"
data:
  - name: "data"
    desc: "2019/12/19"
    chr_prefix: "chr"
    ref_id: "hg38"
    source:
      xg: "static/\*.xg" # To be rewritten
      csv: "static/\*.pcf" # To be rewritten
#      gam: "static/b.gam"
#      gamindex: "static/b.gam.index"
    features: []
    static_files: []

# Specify the version you used to build xg index.
FROM quay.io/vgteam/vg:v1.14.0 as build

# frontend container
FROM momigteam/momig-backend

COPY --from=build /vg/bin/vg /vg/bin/
# Move these files into /vg/static/ folder.
COPY static/\*.xg /vg/static/
COPY static/\*.pcf /vg/static/
COPY static/config.yaml /vg/static/
EXPOSE 8081

CMD ["./graph-genome-browser-backend", "--config=static/config.yaml", "--interval=1500000", "--http=0.0.0.0:8081", "--api=/api/v2/"]

If you use the later VG (version >= 10), gam.index file is no longer used. Please use sorted.gam.gai instead (bam2gam.sh generates gam.gai file).

• static/

  • config.yaml: a configuration file
  • *.xg: an index of a variation graph, generated by [vg](https://github.com/vgteam/vg)
  • *.pcf: a pair of coordinate format file: required to display variants on Feature Table or Circos Plot.
  • *.sorted.gam(optional): read alignments on the graph
  • *.sorted.gam.gai(optional): index of gam

Then, run the MoMI-G backend.

$ docker build -t momig-custom-backend -f Dockerfile.backend .
$ docker run --init -p 8081:8081 -v `pwd`/static:/vg/static momig-custom-backend

At last, run the MoMI-G frontend.

$ sed -e "s/\"target/\"target_/g"  -e "s/\_target/target/g" -i.bak package.json
$ yarn
$ yarn start