Skip to content

Indexing a Reference Genome (L12)

Class Exercise #2: Indexing a Genomic Reference (FASTA)

For this tutorial, we will be using an Organism-specific biological database called Cryptodb. These types of databases are updated more frequently than broader biological databases like Ensembl, especially for non-model organisms. As a result, they tend to be more comprehensive.

If you are working with a non-model organism, it is important to check whether a dedicated database exists for your species, as it may provide more up-to-date and detailed information.

The steps we will perform include:

  1. Making a directory called indexed_genomes_example
  2. Grab the correct FASTA file from the Cryptodb website
  3. Load the hisat2 program using module load
  4. Build the genome index using hisat2-build

What is the purpose of building a genome index?

Prior to mapping, most aligners require you to construct and index the genome, so that the aligner can quickly and efficiently retrieve reference sequence information.

Indexing in general is widely used in bioinformatics to improve performance. This step can take a long time depending on how large the genome is (i.e. hours) so it is a good idea to generate a script.

For this example, we will be indexing the genome of the parasite Cryptosporidium parvum, which is very small in size.

Instructions

  1. Make a directory called indexed_genomes_example

  2. Grab the correct FASTA file from the Cryptodb website and place inside of indexed_genomes_example. 

    https://cryptodb.org/common/downloads/Current_Release/CparvumIOWA-ATCC/fasta/data/
    • Which file will you select? Think about the TYPE of aligner we are using.
    • Once you make a decision, right-click, select copy-link

    Use the wget command 

    wget path-you-copied

    You should see the following: 

    Connecting to cryptodb.org (cryptodb.org)|128.192.21.13|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 9275167 (8.8M) [application/x-fasta]
Saving to: ‘CryptoDB-67_CparvumIOWA-ATCC_Genome.fasta’

100%[=================================================================>] 9,275,167   5.93MB/s   in 1.5s

    Check the file with head once the download is complete. 

    >CP044422 | organism=Cryptosporidium_parvum_IOWA-ATCC | version=2022-10-21 | length=920510 | SO=chromosome
AAACCCCTAAACCTAAACCTAAACCTAAACCTAAACCCTAAACCTAAACCTAAACCTAAA
CCTAAAACCTAAACCTAAAACCTAAACCTAAAACCTAAAACCTAAACCTAAACCTAAACC
CCTAAACCTAAACCTAAACCTAAAAACCTAAACCCTAAACCTAAACCTAAAAAACCTAAA
CCTAAACCTAAACCTAAACCTAAACCTAAACCTAAACCTAAAAAACCTAAAACCTAAACC
TAAACCTAAACCTAAACCTAAAACCTAAAAACCTAAACCTAAAAACCTAAAAAACCTAAA
CCTAAAAAACCTAAACCTAAACCTAAAACCTAAAACCTAAACCTAAAACCTAAAACCTAA
ACCTAAACTAAACCTAAAAAACCTAAAAACCTAAACCCCTAAACCTAAACCTAAAACCTA
AACCTAAACCTAAACCTAAACCTAAAACCTAAACCTAAACCTAAAACCTAAACCTAAACC        
TAAACCTAAACCTCCTAAAAACCTAAACCTAAACCTAAACCTAAACCTAAAAAACCTAAA

  3. Load the hisat2 program using module load 

    module load gcc/13.3.0-xp3epyt 
module load hisat2/2.2.1-x7h4grf

  4. Build the genome index using the hisat2-build program. hisat2-build can index reference genomes of any size. 

    hisat2-build

    Output: 

    Usage: hisat2-build [options]* <reference_in> <ht2_index_base>
    reference_in            comma-separated list of files with ref sequences
    hisat2_index_base       write ht2 data to files with this dir/basename
Options:
    -c                      reference sequences given on cmd line (as
                            <reference_in>)

    To index a reference genome using hisat2-build, the basic command syntax is:

    hisat2-build input-fasta-file basename
    • input-fasta-file: The reference genome FASTA file you want to index.
    • basename: The prefix for the output index files. This is the "name" that will be used for all generated index files.

    For this case, use CparvumIOWA_ATCC_v68 as the basename

Final Ouput

Small indexes are stored in files with the .ht2 extension. These files together constitute the index: they are all that is needed to align reads to that reference. The original sequence FASTA files is no longer used by HISAT2 once the index is built.

The final output should look similar to below once complete: 

CparvumIOWA_ATCC_v68.1.ht2  CparvumIOWA_ATCC_v68.6.ht2
CparvumIOWA_ATCC_v68.2.ht2  CparvumIOWA_ATCC_v68.7.ht2
CparvumIOWA_ATCC_v68.3.ht2  CparvumIOWA_ATCC_v68.8.ht2
CparvumIOWA_ATCC_v68.4.ht2  
CparvumIOWA_ATCC_v68.5.ht2

Check that the file sizes are correct:

-rw-r--r-- 1 pdrodrig pi-jdragon 7.0M Feb 19 11:45 CparvumIOWA_ATCC_v68.1.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon 2.2M Feb 19 11:45 CparvumIOWA_ATCC_v68.2.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon   80 Feb 19 11:45 CparvumIOWA_ATCC_v68.3.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon 2.2M Feb 19 11:45 CparvumIOWA_ATCC_v68.4.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon 3.9M Feb 19 11:45 CparvumIOWA_ATCC_v68.5.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon 2.3M Feb 19 11:45 CparvumIOWA_ATCC_v68.6.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon   12 Feb 19 11:45 CparvumIOWA_ATCC_v68.7.ht2
-rw-r--r-- 1 pdrodrig pi-jdragon    8 Feb 19 11:45 CparvumIOWA_ATCC_v68.8.ht2