Skip to content

Latest commit

 

History

History
184 lines (159 loc) · 13.6 KB

README.md

File metadata and controls

184 lines (159 loc) · 13.6 KB
Raw

IDeRare

IDeRare or "Indonesia Exome Rare Disease Variant Discovery Pipeline" is simple and ready to use variant discovery pipeline to discover rare disease variants from exome sequencing data.

Authored by

Ivan William Harsonoa, Yulia Arianib, Beben Benyaminc,d,e, Fadilah Fadilahf,g, Dwi Ari Pujiantob, Cut Nurul Hafifahh

aDoctoral Program in Biomedical Sciences, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia.
bDepartment of Medical Biology, Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia.
cAustralian Centre for Precision Health, University of South Australia, Adelaide, SA, 5000, Australia.
dUniSA Allied Health and Human Performance, University of South Australia, Adelaide, SA, 5000, Australia.
eSouth Australian Health and Medical Research Institute (SAHMRI), University of South Australia, Adelaide, SA, 5000, Australia.
fDepartment of Medical Chemistry, Faculty of Medicine, Universitas Indonesia, Jalan Salemba Raya number 4, Jakarta, 10430, Indonesia.
gBioinformatics Core Facilities - IMERI, Faculty of Medicine, Universitas Indonesia, Jalan Salemba Raya number 6, Jakarta, 10430, Indonesia .
hDepartment of Child Health, Dr. Cipto Mangunkusumo Hospital, Faculty of Medicine, University of Indonesia, Jakarta, Indonesia.

How to Cite

Please kindly cite the main paper titled "IDeRare: a lightweight and extensible open-source phenotype and exome analysis pipeline for germline rare disease diagnosis" available at https://doi.org/10.1093/jamiaopen/ooae052

Example :

Ivan William Harsono, Yulia Ariani, Beben Benyamin, Fadilah Fadilah, Dwi Ari Pujianto, Cut Nurul Hafifah, IDeRare: a lightweight and extensible open-source phenotype and exome analysis pipeline for germline rare disease diagnosis, JAMIA Open, Volume 7, Issue 2, July 2024, ooae052, https://doi.org/10.1093/jamiaopen/ooae052

Description

  • IDeRare split into 2 parts, namely phenotype and genotype part.
  • This pipeline is designed to be used in Linux environment
  • Original paper may used different version of tools, and the prerequisite used in this pipeline is the latest version of the tools
  • This pipeline is designed and tested with Indonesia rare disease trio patient, but it should be also usable for general cases of rare disease variant discovery from Exome Sequences data given paired end .fq.gz file and HPO data(s)
  • Ensure you have at least 250GB free for database and application setup, and 100GB free for each Trio family exome set
  • The .yaml file path are assuming all the folder are stored in Downloads folder with subfolder of Database (for RefSeq, dbNSFP, dbSNP, ClinVar), Sandbox (for application and its database), IDeRare (git cloned folder)

Data Example

Quick Installation

  1. Clone this repository
git clone https://github.com/ivanwilliammd/IDeRare
  1. Have a Linux environment (Ubuntu or Ubuntu-like 22.04 LTS distro is recommended)
  2. Install Docker and Anaconda - optional- see Prerequisite.md for more details
  3. Run dependency installation script and database script
cd installation
source install_dependencies.sh
source download_database.sh
cd ../
  1. Optional : if you have difficulty in downoading the database and installing the executable dependencies, you could download the data from here and extract it to the Downloads folder, the run the script below to execute install_dependencies.sh from step 7
source install_dependencies.sh --skip-executable

Tips :

To log all bash script (.sh) run, you could use script command to log the terminal output to a file. Example : script -a install.log

IDeRare Phenotype Pipeline - Phenotype Translation, Linkage Analysis, Phenotype Similarity Scoring, Gene-disease recommendation (iderare_phenomizing.py)

Interactive Webapps Implementation of iderare_phenomizing.py hosted at Streamlit and based on our homemade Python library iderare-pheno Python library

IDeRare-Pheno stat : PyPI License Streamlit

  1. This script is recommended if you would like to do conversion, linkage analysis, similarity scoring, and gene-disease recommendation based on the phenotype data provided at clinical_data.txt. Full feature :
    1. Convert the phenotype data to HPO code (accept mixed SNOMED, LOINC, and HPO code)
    2. Similarity scoring of differential diagnosis
    3. Linkage analysis of differential diagnosis (accept mixed SNOMED, ICD-10, ORPHA, OMIM code), include dendrogram tree visualization.
      • This should help clinician to systematically doing work-up and excluding similar diagnosis together based on the patient's phenotype.
    4. Gene and disease recommendation based on the phenotype data similarity scoring between phenotype and OMIM gene and disease databank.
    5. Linkage analysis of recommended causative gene and disease based on phenotype data (include dendrogram tree visualization).
      • This should help clinician to explore / enrich their differential diagnosis based on the patient's phenotype.
    6. Example of the clinical data provided at Clinical Information Example section
  2. Run iderare_phenomizing.sh
  3. Advance usage : you could add extra parameter of threshold, top-n differential diagnoses gene, and number of gene / diagnoses recommendation by using the
# Threshold : float 0 to 1 - default : 0.4 . Description : Threshold of similarity score to be considered as similar
# Differential : int - default : 10 . Description : Top-n differential diagnosis to be furtherly focused on linkage analysis / dendrogram plot, used only if there are no diagnoses passing the minimum threshold
# Recommendation : int - default : 20 . Description : Top-n gene / disease recommendation to be furtherly focused on linkage analysis / dendrogram plot, used only if there are no diagnoses passing the minimum threshold
### Note : all tsv file in output contained all differential diagnoses, and all gene/disease recommendation with their respective similarity score

source iderare_phenomizing.sh --threshold 0.5 --differential 5 --recommendation 5

For more detail, please refer to iderare-pheno Playbook for demonstration of phenotype conversion and analysis.

  1. The output of this file will be saved on output folder, with the file tree and explanation as following.
.
└── output
    ├── {datetime}_linkage_all_ddx.png
    ├── {datetime}_linkage_filt_ddx.png
    ├── {datetime}_linkage_gene.png
    ├── {datetime}_linkage_disease.png
    ├── {datetime}_differential_diagnosis_similarity.tsv
    ├── {datetime}_recommended_disease_similarity.tsv
    ├── {datetime}_recommended_gene_similarity.tsv
    ├── {datetime}_transformed_hpo_set.tsv
    ├── {datetime}_transformed_omim_set.tsv
    └── {datetime}_transformed_hpo_set.txt
File name Description
{datetime}_Linkage of DDx.png dendrogram of the linkage analysis of DDx provided on clinical_data.txt (all)
{datetime}_Linkage of DDx with threshold .png dendrogram of the linkage analysis of DDx provided on clinical_data.txt (threshold)
{datetime}_Linkage of Causative Gene with.png dendrogram of potential causative top-n candidate gene related to patient's phenotype (from HPO OMIM database)
{datetime}_Linkage of Causative Disease w.png dendrogram of potential causative top-n candidate disease related to patient's phenotype (from HPO OMIM database)
{datetime}_differential_diagnosis_similarity.tsv TSV file of differential diagnosis similarity score
{datetime}_differential_recommended_disease_similarity.tsv TSV file of all disease similarity score
{datetime}_differential_recommended_gene_similarity.tsv TSV file of all gene similarity score
{datetime}_transformed_hpo_set.tsv Converted clinical_data to readily used HPO code
{datetime}_transformed_hpo_set.tsv Converted clinical_data to readily used HPO list for yml

IDeRare Genotype Pipeline - Preparing the iderare.yml for exome analysis and phenotype-based variant prioritization

  1. Set the data, directory file reference and trio information on iderare.yml.

    Note : all exome files should be located in the input/A_FASTQ folder of absolute path setup by data_dir at iderare.yml. Example of filled yml available on example/iderare_example.yml

    File Structure Example File
  2. Run the bash script
# Mode : both / solo / trio - default : both . Both will run the pipeline for solo and trio exome data analysis.
# Trimming : true / false - default : false . True will run the trimming process using fastp

source iderare.sh --mode solo --trimming false

Appendix

Clinical Information Example

  • Coded clinical information example in txt format provided at example/clinical_data_example.txt.
  • This clinical information is the patient phenotype and differential diagnoses complementing trio exome data provided at Bioproject database 1077459

Phenotype Data

Clinical Finding Source of Information Coded in EMR Code Interpretation Writing format at clinical_data.txt
Autosomal recessive inheritance Inheritance Pattern SNOMED-CT 258211005 SNOMEDCT:258211005
Hepatosplenomegaly Physical Examination SNOMED-CT 36760000 SNOMEDCT:36760000
Anemia Physical Examination SNOMED-CT 271737000 SNOMEDCT:271737000
Ascites Physical Examination SNOMED-CT 389026000 SNOMEDCT:389026000
Inadequate RBC production Problem List SNOMED-CT 70730006 SNOMEDCT:70730006
Abnormality of bone marrow cell morphology Problem List SNOMED-CT 127035006 SNOMEDCT:127035006
Cholestasis Problem List SNOMED-CT 33688009 SNOMEDCT:33688009
Abnormal liver function Problem List SNOMED-CT 75183008 SNOMEDCT:75183008
Impending hepatic failure Problem List SNOMED-CT 59927004 SNOMEDCT:59927004
Osteopenia Problem List (Radiology Finding) SNOMED-CT 312894000 SNOMEDCT:312894000
Mitral regurgitation Problem List (Cardiology Finding) SNOMED-CT 48724000 SNOMEDCT:48724000
Metabolic alkalosis Problem List (Blood Gas Analysis) SNOMED-CT 1388004 SNOMEDCT:1388004
Low Albumin Serum Level Clinical Pathology (Lab) LOINC 1751-7 L LOINC:1751-7
Low HDL Level Clinical Pathology (Lab) LOINC 2085-9 L LOINC:2085-9
Low Platelet Count Clinical Pathology (Lab) LOINC 777-3 L LOINC:777-3
Increased Lactate Level Clinical Pathology (Lab) LOINC 2519-7 H LOINC:2519-7
Increased ALT Level Clinical Pathology (Lab) LOINC 1742-6 H LOINC:1742-6
Increased AST Level Clinical Pathology (Lab) LOINC 1920-8 H LOINC:1920-8
Abnormal lower motor neuron Disease Spectrum related to EMG result HPO 0002366 HP:0002366
Increase Hepatic Glycogen Content Liver Biopsy Pathology Interpretation HPO 0006568 HP:0006568
Bone-marrow foam cells Pathology Anatomy Bone Marrow Aspiration HPO 0004333 HP:0004333
Failure to thrive during infancy Developmental history HPO 0001531 HP:0001531

Working diagnosis before Exome Sequencing

Differential Diagnosis Code Type EMR Code Writing format at clinical_data.txt
Beta thalassemia SNOMED-CT 65959000 SNOMEDCT:65959000
Gaucher Disease SNOMED-CT 190794006 SNOMEDCT:190794006
Niemann Pick Disease type C SNOMED-CT 66751000 SNOMEDCT:66751000
Glycogen storage diseases spectrum ICD10 E74.0 ICD-10:E74.0