stat:bioinform:majiq

/mnt/smb/notebooktian/RNASeq/260429AGING

### References * MAJIQ [RNA splicing analysis using heterogeneous and large RNA-seq datasets - nature communications](https://www.nature.com/articles/s41467-023-36585-y) * MAJIQ → cryptic 3'splice site

* A3SS 图B，有拖尾感的sashimi [Cancer-Associated SF3B1 Hotspot Mutations Induce Cryptic 3′ Splice Site Selection through Use of a Different Branch Point](https://www.cell.com/cell-reports/fulltext/S2211-1247(15)01078-5)

### MAJIQ

https://bitbucket.org/biociphers/majiq_academic/src/main/

https://biociphers.bitbucket.io/majiq-docs/

```dockerfile FROM python:3.12-slim-bookworm

ENV DEBIAN_FRONTEND=noninteractive

# 1. 安装系统依赖（已增加 libzstd-dev） RUN apt-get update && apt-get install -y –no-install-recommends \

  build-essential \
  git \
  libhts-dev \
  libbz2-dev \
  zlib1g-dev \
  liblzma-dev \
  libzstd-dev \
  && rm -rf /var/lib/apt/lists/*

ENV HTSLIB_LIBRARY_DIR=/usr/lib/x86_64-linux-gnu ENV HTSLIB_INCLUDE_DIR=/usr/include

# 2. 拆分指令以利用缓存 (存档点：下载源码) WORKDIR /tmp RUN git clone https://bitbucket.org/biociphers/majiq_academic.git

# 3. 存档点：安装 Voila WORKDIR /tmp/majiq_academic RUN pip install –no-cache-dir ./moccasin RUN pip install –no-cache-dir ./voila

# 4. 存档点：安装 MAJIQ RUN pip install –no-cache-dir ./majiq

WORKDIR / RUN rm -rf /tmp/majiq_academic

RUN majiq –help && voila –version

CMD [“/bin/bash”] ```

#### License * 放入 ~/majiq_license_academic_official.lic ```majiq_license_academic_official.lic gAAAAABlW6-jYooOgeYvzqoEZ9Z6jMLcFAfydEzRxJWwJ7bl9z71xmQCS343USkuLpu88lpTwTOGiGqpJ0uMZV36nto1SCRmdP_LWSugZyS7-E1wDVOzyV8kQcUHzNtPQVMVaLUQXmpYGfXGHEVo-XUPVG3v3waurw== ```

### Steps v3 ```bash alias majiq='docker run –rm -e MAJIQ_LICENSE_FILE=/data/majiq_license_academic_official.lic -v “$(pwd)/datamj:/data” -w /data majiq_image majiq' alias voila='docker run –rm -e MAJIQ_LICENSE_FILE=/data/majiq_license_academic_official.lic -v “$(pwd)/datamj:/data” -p 5000:5000 -w /data majiq_image voila' alias majiq3='docker run –rm -e MAJIQ_LICENSE_FILE=/data/majiq_license_academic_official.lic -v “$(pwd)/datamj:/data” -w /data majiq_image majiq-v3'

PATH_GFF3=“/data/mm39_115.gff3” PATH_BAMS=“/data/bams” PATH_ANNO_SG=“/data/mj/anno_sg.zarr” PATH_SJ=“/data/mj/sjs” PATH_SG_ZAAR=“/data/mj/sg.zarr” PATH_PSICOV=“/data/mj/psi” PATH_SGC=“/data/mj/sgc” PATH_VOILA=“/data/mj/voila”

majiq3 gff3 $PATH_GFF3 $PATH_ANNO_SG majiq3 sj $PATH_BAMS/OLDSED_1.bam $PATH_ANNO_SG $PATH_SJ/OLDSED_1.sj majiq3 sj $PATH_BAMS/OLDSED_2.bam $PATH_ANNO_SG $PATH_SJ/OLDSED_2.sj majiq3 sj $PATH_BAMS/OLDSED_3.bam $PATH_ANNO_SG $PATH_SJ/OLDSED_3.sj majiq3 sj $PATH_BAMS/YNGSED_1.bam $PATH_ANNO_SG $PATH_SJ/YNGSED_1.sj majiq3 sj $PATH_BAMS/YNGSED_2.bam $PATH_ANNO_SG $PATH_SJ/YNGSED_2.sj majiq3 sj $PATH_BAMS/YNGSED_3.bam $PATH_ANNO_SG $PATH_SJ/YNGSED_3.sj

majiq3 build $PATH_ANNO_SG $PATH_SG_ZAAR –sjs $PATH_SJ/OLDSED_1.sj $PATH_SJ/OLDSED_2.sj $PATH_SJ/OLDSED_3.sj –sjs $PATH_SJ/YNGSED_1.sj $PATH_SJ/YNGSED_2.sj $PATH_SJ/YNGSED_3.sj

majiq3 psi-coverage $PATH_SG_ZAAR $PATH_PSICOV/OLDSED.psicov $PATH_SJ/OLDSED_1.sj $PATH_SJ/OLDSED_2.sj $PATH_SJ/OLDSED_3.sj majiq3 psi-coverage $PATH_SG_ZAAR $PATH_PSICOV/YNGSED.psicov $PATH_SJ/YNGSED_1.sj $PATH_SJ/YNGSED_2.sj $PATH_SJ/YNGSED_3.sj

majiq3 sg-coverage $PATH_SG_ZAAR $PATH_SGC/OLDSED.sgc $PATH_SJ/OLDSED_1.sj $PATH_SJ/OLDSED_2.sj $PATH_SJ/OLDSED_3.sj majiq3 sg-coverage $PATH_SG_ZAAR $PATH_SGC/YNGSED.sgc $PATH_SJ/YNGSED_1.sj $PATH_SJ/YNGSED_2.sj $PATH_SJ/YNGSED_3.sj

majiq3 deltapsi –splicegraph $PATH_SG_ZAAR –output-voila $PATH_PSICOV/OLDSED_v_YNGSED.psicov –output-tsv $PATH_PSICOV/OLDSED_v_YNGSED.tsv -psi1 $PATH_PSICOV/OLDSED.psicov -psi2 $PATH_PSICOV/YNGSED.psicov

majiq3 quantify $PATH_PSICOV/OLDSED.psicov –min-experiments 0.01 –splicegraph $PATH_SG_ZAAR –output-tsv $PATH_PSICOV/OLDSED.tsv –overwrite majiq3 quantify $PATH_PSICOV/YNGSED.psicov –min-experiments 0.01 –splicegraph $PATH_SG_ZAAR –output-tsv $PATH_PSICOV/YNGSED.tsv –overwrite

majiq3 heterogen –stats infoscore mannwhitneyu ttest tnom –splicegraph $PATH_SG_ZAAR –output-voila $PATH_VOILA/OLDSED_v_YNGSED.voila –output-tsv $PATH_PSICOV/OLDSED_v_YNGSED.tsv -psi1 $PATH_PSICOV/OLDSED.psicov -psi2 $PATH_PSICOV/YNGSED.psicov

voila view –host 0.0.0.0 –port 5000 –enable-type-indexing –psicov-grouping-file /data/majiq.tsv $PATH_SG_ZAAR $PATH_VOILA/OLDSED_v_YNGSED.voila $PATH_SGC/

```

### Steps v2 ```bash majiq build /data/mm39_115.gff3 -c /data/majiq.config -o /data/build_output majiq psi-coverage \

  /data/build_output/splicegraph.zarr \
  /data/psi_coverage_output \
  /data/build_output/YNGSED_1.sj /data/build_output/YNGSED_2.sj /data/build_output/YNGSED_3.sj \
  /data/build_output/OLDSED_1.sj /data/build_output/OLDSED_2.sj /data/build_output/OLDSED_3.sj \
  --nthreads 8

majiq deltapsi \

grp1 /data/psi_coverage_output \
grp2 /data/psi_coverage_output \
-select-grp1-prefixes YNGSED_1 YNGSED_2 YNGSED_3 \
-select-grp2-prefixes OLDSED_1 OLDSED_2 OLDSED_3 \
-names YNGSED OLDSED \
-splicegraph /data/build_output/splicegraph.zarr \
-output-voila /data/deltapsi_output/YNGSED_vs_OLDSED.voila \
-output-tsv /data/deltapsi_output/YNGSED_vs_OLDSED.tsv \
-nthreads 8

majiq sg-coverage /data/build_output/splicegraph.zarr /sgc/YNGSED.sgc /data/build_output/YNGSED_1.sj /data/build_output/YNGSED_2.sj /data/build_output/YNGSED_3.sj

voila view /data/build_output/ –host 0.0.0.0 –port 5000 –enable-type-indexing –psicov-grouping-file /data/majiq.tsv

```