Skip to content

02_Guide_to_Build_SnpEff_DB.md

Latest commit

 

History

History
149 lines (95 loc) · 6.91 KB

02_Guide_to_Build_SnpEff_DB.md

File metadata and controls

149 lines (95 loc) · 6.91 KB

Building a Custom SnpEff Database

Table of Contents

  1. Introduction
  2. Downloading Reference and Annotation Data
  3. Setting Paths and Navigating to the snpEff Directory
  4. Editing snpEff.config
  5. Building the SnpEff Database
  6. Handling Errors During Build
  7. Verifying Mutations with IGV

1. Introduction

In this tutorial, we will guide you through the process of building a custom SnpEff database using NCBI RefSeq Annotations for rice (Oryza sativa). The steps outlined here are meant for users who need to annotate genomic variants using their own reference genome or custom annotations that may not be available in the pre-built SnpEff databases.

Before proceeding, it is highly recommended to carefully read the SnpEff documentation. This tutorial assumes that you have already installed SnpEff using Conda (conda install -c bioconda snpeff). If your reference genome is already available in a pre-built SnpEff database, you can avoid building a custom database. For MutMap users, please refer to the MutMap manual for details on how to use the -e option to select pre-built databases.

2. Downloading Reference and Annotation Data

For this tutorial, we will use the NCBI RefSeq Annotations for rice (Oryza sativa). The dataset can be accessed at the following URL:

NCBI RefSeq Annotations for Oryza sativa

We will use the following links to download the reference FASTA and GFF files using wget, then decompress them using gunzip:

To download and decompress these files, run the following commands:

# Download reference FASTA
wget https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/034/140/825/GCF_034140825.1_ASM3414082v1/GCF_034140825.1_ASM3414082v1_genomic.fna.gz

# Download GFF file
wget https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/034/140/825/GCF_034140825.1_ASM3414082v1/GCF_034140825.1_ASM3414082v1_genomic.gff.gz

# Decompress the files
gunzip GCF_034140825.1_ASM3414082v1_genomic.fna.gz
gunzip GCF_034140825.1_ASM3414082v1_genomic.gff.gz

Next, define the paths to these files as variables:

PATH_TO_FASTA=$(pwd)/GCF_034140825.1_ASM3414082v1_genomic.fna
PATH_TO_GFF=$(pwd)/GCF_034140825.1_ASM3414082v1_genomic.gff

3. Setting Paths and Navigating to the snpEff Directory

Once the reference and annotation files are downloaded and paths are set, you need to navigate to the directory where snpEff is installed. If you installed snpEff using Conda, it is typically located at:

${HOME}/anaconda/envs/mutmap/share/snpeff-5.2-1

However, the exact path may vary depending on whether you are using Conda or Mamba, and it may also differ based on the version of snpEff you have installed.

After navigating to this directory, you should see the following directories and files:

  • scripts
  • snpEff
  • snpEff.config
  • snpEff.jar

Now, create a directory for your custom genome data:

mkdir -p data/GCF_034140825.1

Next, create symbolic links for the reference FASTA and GFF files in the newly created directory:

ln -s ${PATH_TO_FASTA} data/GCF_034140825.1/sequences.fa
ln -s ${PATH_TO_GFF} data/GCF_034140825.1/genes.gff

4. Editing snpEff.config

After setting up the files, the next step is to edit the snpEff.config file to register your custom genome. Open snpEff.config in a text editor and add the following line:

GCF_034140825.1.genome : GCF_034140825.1

Make sure that the second GCF_034140825.1 matches the directory name you created in the data/ folder. This line tells snpEff where to find the genome data for your custom annotation.

Save the file after adding the line.

5. Building the SnpEff Database

Once the snpEff.config file is updated, you can start building the SnpEff database using the following command:

snpEff build -gff3 -noCheckCds -noCheckProtein GCF_034140825.1

This command builds the database using the reference genome and annotation files located in data/GCF_034140825.1/.

6. Handling Errors During Build

If the GFF file is correctly formatted and compatible with snpEff, the build process will complete successfully, and you will be able to annotate variants using the -e option with the specified genome.

However, in some cases, you might encounter errors such as:

WARNING_CHROMOSOME_CIRCULAR: Chromosome 'NC_001320.1'
WARNING_GENE_NOT_FOUND: Gene 'null' (NC_001751.1:1732-2019)

These indicate that the build process has failed due to issues with annotations for certain contigs (e.g., organelle sequences) in the GFF file.

Removing Problematic Contigs from the GFF

To fix this issue, you can remove the problematic contig annotations from the GFF file before building the database. Below is an example of how to do this using grep to filter out both NC_001320.1 and NC_001751.1:

# Create a new GFF file excluding the problematic contigs
grep -v -e "NC_001320.1" -e "NC_001751.1" ${PATH_TO_GFF} > filtered_genes.gff

# Update the symbolic link to point to the new filtered GFF file
ln -sf $(pwd)/filtered_genes.gff data/GCF_034140825.1/genes.gff

This command will exclude any lines in the GFF file that contain NC_001320.1 or NC_001751.1, the contigs causing the issue. After creating the new filtered_genes.gff file, update the symbolic link for the GFF file in the snpEff data directory.

Re-running the Build Command

Once the GFF file has been filtered, you can re-run the snpEff build command:

snpEff build -gff3 -noCheckCds -noCheckProtein GCF_034140825.1

If the problematic contigs were the only cause of the error, the build should now complete successfully.

7. Verifying Mutations with IGV

Even if your SnpEff database is built without errors, it is important to manually verify the annotations. GFF format inconsistencies may still cause SnpEff to annotate variants incorrectly, even when no errors are reported during the build process.

To check the accuracy of the annotations, you can use a genome visualization tool like IGV (Integrative Genomics Viewer). IGV allows you to visually inspect the variant locations and assess their impact on gene function.

You can download IGV and learn more about it from the following link:

IGV