Skip to content

How to run BioNanoAnalyst

Jump to bottom
Andy Yuan edited this page Apr 27, 2017 · 27 revisions

This is a tutorial page helping users to quickly get familiar with BioNanoAnalyst.

Steps:

  1. Select a reference genome in fasta/fa/fna format (required).

  2. If using raw BioNano maps (Option 1), please ensure "scripts", "Assembler" and " RefAligner" from BioNano Genomics are available in your system (see below for more detailed instructions). For this option, a raw BioNano bnx file obtained from BioNano Genomics platform, such as Irys platform is required. After selecting a proper enzyme, importing the bnx file and setting the parameters under 'Settings', the assembly and alignment can be started by clicking the 'Start' button. Note: Option 1 is currently not available in Windows system. When running this step, it can take a while depended on the input files and the settings. We recommend to run this step on a cluster if you have a big dataset (>100Mb). For your convenience, we have also provided a script to help run jobs on a cluster: https://github.com/AppliedBioinformatics/runBNG.

  3. If using aligned BioNano data (Option 2), an xmap file and the corresponding ref and query cmap files are needed. Note: the aligned files are generated by comparing a ref and qry maps, such as EXP_REFINAL1.cmap from the /exp_refineFinal1 folder as the query map and in silico digested NGS sequences as a ref map.They are not the de novo assembled sub-files.

  4. In the analysis step, a confidence score should be specified (≥0) before starting. Bioinformatician from BioNano Genomics suggests using 15 as a good start. Usually, we recommend a confidence score between 10-20, however, users can adjust this number depending on their own mapping quality. Below is a table interpreting a confidence score and its association with the quality of nucleotide mappings from the comparison between human NGS assembly data: hg19_chr1 and all contigs from 24/05/2000 assembly which were used to build chr1. The method was to digest hg19_chr1 and all contigs in silico using BspQI to get cmap files and then use hg19_chr1.cmap as a ref cmap and ctgs.cmap as a query map to map them using RefAligner. After getting an xmap file, we used blastn to align each mapped contig with chr1 in their mapped regions reported in the xmap file under a corresponding confidence score. In blastn result, we only selected the top hit with an identity rate>=99%, Evalue ~0. Note: The table below provides a tested result for users to check the mapping stats with a corresponding confidence score. To decided which confidence score should be used, users may visually check the mapping quality or check the xmap file. The tricky thing is if using a large confidence score, information below this selected confidence score will be hidden, which means users may lose many mappings. A proper confidence score is really depended on the research purpose that users want to achieve.

  Overall:
  Mapping: ctgs_chr1_to_hg19_chr1
  Min_confidence_score: 12.25
  Max_confidence_score: 544.78
  Total_number_of_mapping: 109
  Total_refer_mapped_length: 95146336 bp

  confidence_score average_mapping_size(bp)*     confidence_score  map_left(%)  total_ref_mapped_len (bp)
   [10, 15)         66393                         >=10              100.00       95146336
   [15,20)          70402                         >=15              90.83        93835874
   [20,25)          102637                        >=20              77.98        91670481
   [25,30)          111613                        >=25              69.72        90008858
   [30,35)          194000                        >=30              57.80        86644558
   [35,40)          197171                        >=35              50.46        84090226
   [40,45)          167396                        >=40              45.87        82112264
   [45,50)          202646                        >=45              41.28        79622104
   [50,55)          240851                        >=50              38.53        78516942
   [55,60)          188613                        >=55              33.94        75394112
    >=60            235605                        >=60              32.11        74036462
   
   *Result from BLASTN best/top hit

  1. After specifying a confidence score, click the "Analyse" button, BioNanoAnalyst will run and the "Job status" panel will immediately show the job status as "Running" in blue colour or "Finished" in green colour. Note: When BioNanoAnalyst is running, it may temporarily stop responding due to the heavy computing load. If this occurs, please let BioNanoAnalyst run and do not perform any action within the program.

  2. After the "Job status" indicates "Finished", the tabs in the top right panel show the results of mapping for each stop of the workflow, as a table of summary statistics and as a pie chart. Plots of the mapping results across all sites analysed is available via the bottom panel.

  3. To view the summary mapping plot for a contig of interest, select a contig from the drop-down menu in the "Mapping Plot" panel. Users can zoom the plot and click specific site to check its site ID and position on the reference. A copy of the plot shown can be saved to a specified directory; alternatively all contig plots can be saved using the "Save all plots" button. Users may also save a gff3 file into a selected directory, which contains all information reported by BioNanoAnalyst by clicking the "save quality report" button.

Setting up an analysis using raw BioNano data:

  1. BioNanoAnalyst provides two options for analysing BioNano data in Linux/Unix system and macOS system. If users want to use raw BioNano data (Option 1), please download the BioNano tools "Assembler", "RefAligner" and the "scripts" folder from BioNano Genomics. RefAligner and Assembler can be downloaded at: http://www.bnxinstall.com/RefAlignerAssembler; Scripts can be downloaded at: http://www.bnxinstall.com/Scripts
  2. Install or decompress "Assembler" and "RefAligner" on your local computer and move them into a single folder, making sure the applications are not renamed and are executable. The folder containing these applications will need to be indicated under Option 1 -> Settings -> Tools for the analysis to run.
Clone this wiki locally