MirAge
About
MirAge is a fast analysing tool which is sensitive and highly precise. You can use any reference database build yourself.
It is capable of detecting Bacteria, Virus or any other DNA or RNA species. Even plain text analysis is possible to be used in i.e. plagiarism detection.

Configuration is intuitive, versitile and easy. High memory usage can be set if available, to speed up analysis.
MirAge can be used on your home PC, laptop or even a powerfull HPC (High Performance Computing) environment.
Especially for HPC usage, time estimation of the analysis is added so you can easily obtain an estimated required time to request a time window on your HPC environment.

Please do not hessitate to contact us at mirage@mirdesign.nl.
Results
A) Results of analyzing CAMI 1 study samples
B) Results of performance tests
C) Reported and rerun results of analyzing 1 simulated HiSeq sample from Kraken
D) Analysis results of 7 simulated samples (higher is underlined).
Download MirAge
The same source code can be used to compile for Windows or Unix.
MirAge can be download using GitHub.
Quickstart
Download and Install
  • Unix
  • Open a shell.
    • cd ~
    • mkdir mirage
    • cd mirage
    • git clone https://github.com/mirdesign/mirage.git
    • sh ./compile.sh
  • Windows
  • Open a command prompt.
    • cd c:\
    • md mirage
    • cd mirage
    • git clone https://github.com/mirdesign/mirage.git
    • compile.bat
    • Run

    This will also build a mhm file which contains a quick lookup database. You can re-use that database.
    • ./mirage -query ../sample.to.analyse.fasta -reference ../large.references.fasta
    Re-use build database:
    • ./mirage -query ../sample.to.analyse.fasta -reference ../large.references.fasta.mhm
    Use without parameters, using all settings from configuration.cfg. Edit this file to change the query and reference path.
    • ./mirage
    Download NCBI Reference database
    To download and build a reusable NCBI Refseq database, use below unix commands to download and use the NCBI reference database.
    1. Unix
      1. 1. Retrieve complete NCBI refseq database
        • wget https://ftp.ncbi.nlm.nih.gov/genomes/refseq/bacteria/assembly_summary.txt
        • awk -F "\t" '$12=="Complete Genome" && $11=="latest" {print $20}' assembly_summary.txt > list
        • awk 'BEGIN{FS=OFS="/";fs="genomic.fna.gz"}{fd=$0;asm=$10;f=asm"_"fs;print fd,f}' list > ftplist
        • cat ftp.list | parallel -j 8 wget
        • gunzip *_genomic.fna.gz
        • cat *_genomic.fna >> ncbi.refseq.fasta
        • rm *_genomic.fna
    3. Windows
    5. Get NCBI Taxonomic ID information
      • Now we have ncbi.refseq.fasta containing the full complete genome database.
        Per sequence, an accession ID is provided. To later be able to interper the results, taxonomy ID's are required.
      •  
      • 1. Retrieve Taxonomy data
        • Unix
        • wget https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/accession2taxid/nucl_gb.accession2taxid.gz
        • gunzip nucl_gb.accession2taxid.gz
        • Windows
        • curl https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/accession2taxid/nucl_gb.accession2taxid.gz --output nucl_gb.accession2taxid.gz
        • "C:\Program Files (x86)\7-Zip\7z.exe" x nucl_gb.accession2taxid.gz -y
        • del nucl_gb.accession2taxid.gz
      •  
      • 2. Use NCBI refseq database in MirAge
        • ./mirage -reference ncbi.refseq.fasta -NCBITaxID nucl_gb.accession2taxid -query sample.fasta
      •  
      • This will build a reusable ncbi.refseq.fasta.mhm database and analyse sample.fasta. The results provide TaxonomicID's and the provided header with Accession number and record name.
    Configuration file
    Override configuration.cfg

    MirAge contains a configuration.cfg file which is used as primary setting source.
    You can override each setting via the command prompt parameters directly.

    As example, the configuration files contains progress_time_type with the default value 'remaining'. To override, provide the following parameter:
    • mirage.exe --progress_time_type estimate

    Settings explained

    Name Possible values or (default) Description
    QuerySequencesPath /home/username/data/sample1.fasta The query file to be analysed
    Only 1 occurance allowed.
    ReferenceDatabasePath /home/username/data/database.fasta
    /home/username/data/database.mhm     		The reference file to be used
    Multiple occurances allowed, but only for MHM files.
    Hashmap_Load_Hashmap_Percentage 1 to 100 (100) Load a MHM database but randomly only load x% of the data. This is usefull when a quick scan of your sample is required.
    hashmap_load_hashmap_skip_on_failure 0
    1     		End program when failing to load multiple hashmaps
    Skip hashmap when unable to load.
    Hashmap_Load_Reference_Memory_UsageMode
    setting speed memory usage description
    low Slow Low Lazy loading, only load sequence and positioning information on the fly.
    medium Faster Medium Load complete sequences into memory. Lazy load positioning information on the fly.
    high Fast,
    slow start    		 High Load complete sequences into memory. Pre-load positioning information during start.
    extreme Very fast,
    slow start    		 Very high Load complete sequence and positioning information into memory.
    LogThreshold_HighSensitivity 0, 0.1, 0.2, .. to 1 (0) Threshold for high sensitivity log output.
    LogThreshold_HighPrecision 0, 0.1, 0.2, .. to 1 (1) Threshold for high precision log output.
    LogThreshold_Balanced 0, 0.1, 0.2, .. to 1 (0.7) Threshold for balanced log output.
    Hardware_BlockSize_Default_CPU 1 or higher (5000) Number of sequences to be analysed in 1 block. Higher requires more memory but can saturate CPU better.
    Hardware_maxCPUThreads 1 or higher (75000) Limit of number of threads used.
    Hardware_BlockSize_AutoIncrease_Enabled_Default 1
    0     		Enable autoincrease to let MirAge try to optimise the block size in order to fully utilize the hardware.
    Disable autoincrease.
    Hardware_BlockSize_AutoIncrease_IncreasePercentage 100 to 500 (130) When autoincrease is enabled, how fast should it let the blocksize grow.
    UltraFastMode Sensitive
    Fast    		 Mode to analyse fast or sensitive.
    Output_Mode normal
    silent
    results
    resultsNoLog     		Full application log
    No screen output, only log files
    Show results on screen
    Show results on screen, without creating log files
    output_firstresults_only 1
    0     		Only the references with the highest score are logged
    All scores and their related references are logged. Ordered by score.
    progress_time_type remaining
    estimate     		Show time remaining during analysis
    Show total time estimated during analysis
    considerthreshold 1 and higher (32) The threshold for accepting succesive regions. Lower is more sensitive and less precise.
    OutputPath /home/user/output The path where log and multinode files are stored.
    Screen_Resize (Windows only) 0
    1     		Do not resize.
    Resize the command window
    Screen_COLS (Windows only) 1 and higher (160) Set window column size
    Screen_ROWS (Windows only) 1 and higher (60) Set window row size
    Hardware_Device_Queue_MaxSize (Experimental!) 1 or higher (1) Let the CPU do multiple prepare round while analysing a block
    Publication
    MirAge has been published on https://www.biorxiv.org/content/10.64898/2025.12.02.691787v1 (PDF)
    Please citate MirAge as follow:
    M Pater, M D de Jong, MirAge, a novel Matrix Analyzer for fast DNA classification, bioRxiv 2025.12.02.691787, doi: https://doi.org/10.64898/2025.12.02.691787