GeoMx Digital Spatial Profiler Software Updates v2.0
GeoMx Digital Spatial Profiler v2.0 Instrument Software and Data Analysis Suite Update
With release of software version 2.0, GeoMx expands its functionality to include compatibility with NGS readout on Illumina.
The v2.0 release supports the processing and analysis of the Cancer Transcriptome Atlas (CTA), which is a panel of over 1800 RNA targets to spatially profile the tumor, tumor microenvironment, and tumor immune status.
The release is also coupled with the introduction of a new GeoMx NGS Pipeline that automatically processes Illumina FASTQ sequencing files to GeoMx readable digital counts (DCC) for input back into the DSP Data Analysis Suite (DSPDA).
For users running NGS experiments, a separate server off the instrument is required for running DSPDA.
Feature Updates for v2.0:
- Processing and analysis of the Cancer Transcriptome Atlas (CTA), an NGS panel of over 1800 RNA targets.
- Coordination with the GeoMx NGS Pipeline software to automatically process Illumina FASTQ sequencing files to GeoMx readable digital counts (DCC) for input back into the Data Analysis Suite.
- New data analysis visualizations: Pathway Analysis and Correlation Plot.
- Enhanced Data Analysis features, including Filter function on toolbar.
- Coordination with off-instrument DSP Data Analysis software on server.
- Feature improvements to the Administration controls, including the option to configure off-instrument DSP Data Analysis and to enable/disable probe kit configuration files.
- Performance improvements to Scan Workspace including ROI batch import and ROI Report export.
- New authentication and login page.
- Bug fixes, security, and usability improvements.
System Requirements for Data Anaysis Server for NGS
- Windows 2019 Server with Hyper-V enabled
- HTTPS (Port 443) open inbound
- CPU: Intel processor with 64-bit support, 3.40 GHz or faster, 4 cores
- Memory: 64GB DDR4
- OS SSD: 512GB
- Main Storage HD: 2TB
System Requirements for GeoMx NGS Pipeline
The GeoMx NGS Pipeline uses a series of open-source components to process sequencing files from FASTQ to the Digital Count Conversion (DCC) file that is read by the DSP instrument for further analysis. In the first step the raw sequencing files (FASTQ files) are selected for a specific pipeline run. Once selected the reads are processed for quality, the adapters are removed (trimmed) and the paired-end reads are merged resulting in the high quality read. In the third step the reads are aligned to the Readout Tag Sequence-ID (RTS-ID) barcodes. Then PCR duplicates are removed by matching on the Unique Molecular Index (UMI) and the DCC file is generated. The resulting DCC files are presented as a .zip file in a folder which you designate and can then be uploaded into the DSP Control Center for study creation in the DSP Data Analysis Suite.
There are two components to the GeoMx NGS Pipeline, the Graphical User Interface (GUI) and the Pipeline which processes the files. There is an installer specific to each Operating System supported for the GUI and each Operating System supported for the Pipeline.
|Local and Server Install|
|Windows (UI, local)||Windows 10||Intel Core i5-4750 3.20 GHz||16|
|Mac (UI, local)||MacOS Catalina V.10.15.5||Intel Core i5 2.60 GHz||16|
|Linux (Server)||Linux Ubuntu 18.04||AMD Phenom 8650 2.30 MHz||32|
|CLI Only||AWS t2.micro||AWS Linux 2||vCPU 1||16|
|AWS t3.xlarge||AWS Linux||vCPU 4||16|
For smaller experiments sizes, maximum 12 FASTQ files, the GeoMx NGS Pipeline may be run on a local machine and managed through a graphical user interface. The GUI and pipeline components may be installed on a single machine with the following recommended system requirements:
For Apple OS:
- Macbook pro with a CPU at 1.4 GHz processor or better
- 16GB RAM
For Windows OS:
- Intel® Core™ i5-835OU @ 1.70GHZ 1.90 GHz
- 16 GB RAM
- 64-bit Operating System, x64-based processor
For larger experiments sizes or multiple users the GeoMx NGS Pipeline should be installed on a Linux server or in an AWS instance or on AWS Batch. The Linux server can be managed from the graphical user interface which can be installed on Windows or Mac as above. The AWS implementation is best managed via the command line interface (CLI). The minimum system requirements are as follows:
The Linux server requirements are as follows:
- OS: Ubuntu 18 and up
- 32GB RAM
- 300GB storage
- CPU: Intel(R) Xeon(R) Gold 6248 CPU @2.5GHz, hot add enabled
- Open SSH installed and enabled.
The recommended AWS instance is as follows:
- Platform: x64 system, capable of running Linux containers (Latest Linux or Ubuntu OS)
- CPU & RAM: standard Amazon t3.xlarge (4 cores, 16GB RAM)
- Main storage: 20GB
- Extra File storage: At least dataset size x 10
Dataset for Testing GeoMx NGS Pipeline
These sample FASTQ files can be used to test an installation of the GeoMx NGS Pipeline in a customer’s environment. The corresponding DCC files are also provided to match with outputs from the customer pipeline to ensure the pipeline is functioning properly.
This sample dataset contains NGS data for three samples (wells). The three wells represent two collected AOI’s (area of illumination) and one NTC (no-template control) from a GeoMx DSP experiment.
These samples are from a dataset prepared by NanoString using an early access version of the Cancer Transcriptome Atlas (CTA). These samples were chosen for this test dataset as they align with the expected typical Region of Interest (ROI) size and sequencing quality.
The defined ROI is roughly 300µm diameter and was segmented using PanCK. Both samples (wells) represented in this dataset were PanCK positive from two different ROIs.
Sequencing saturation was between 70% and 80% ((1 – raw reads/deduplicated reads) x 100)%.
|FILE ROOT/SAMPLE NAME||SAMPLE INFO||AREA||SEQUENCING SATURATION|
|DSP-1001660005876-A01||No template control||0 µm2||–|
|DSP-1001660005876-A12||PanCK positive auto segment||86118 µm2||77.2%|
|DSP-1001660005876-C01||PanCK positive auto segment||74208 µm2||70.5%|
Each sample has eight FASTQ files- data are from four lanes of a NextSeq run using paired-end- so there are forward and reverse reads X 4 lanes for each sample.
|3sampleAOIs_20200504_DND_config.ini||The configuration file downloaded from the GeoMx DSP instrument. This file is required by the GeoMx NGS Pipeline (DND 1.0) app to process the FASTQ files.|
|Subfolder: /FASTQ/||Folder containing FASTQ sequencing data from Illumina NextSeq. 8 files per sample (forward & reverse x 4 Lanes), 24 FASTQ files total.|
|Subfolder: /Run1DCCs/||Contains Summary of output, DCC zip file for upload to DSP DA, and sub-folder with DCC output from NanoString for comparisons.|
|Subfolder: /Run1DCCs/DCCs/||The DCCs files to compare with customer test run|
|/Run1DCCs/summary.txt||Summary of number of raw, trimmed, stitched and aligned reads for each sample|
|/Run1DCCs/DCC_files_20200813.zip||Sample of the DCC package which would be uploaded back to your GeoMx DSP|
|Subfolder: /Run1DCCs/logs/||Logs from pipeline|
|Subfolder: /Run1DCCs/Count_Tables/||Spreadsheet summaries of deduplication|
Download the FASTQ folder and contents and the config file to a location where your GeoMx NGS Pipeline (DND 1.0) installation can access them. Use “3sampleAOIs_20200504_DND_config.ini” as the configuration file and the FASTQ folder as the input folder. When your run is complete compare the DCC files in your output folder to those in the provided “Run1DCCs/DCCs” folder. You can inspect DCC file contents with a text editor. Sort order of the <Code_Summary> section may differ between your run and the provided DCCs.
You can copy <Code_Summary> sections to a new file and save as a comma separate value file format to compare in a spreadsheet program
The run should complete with warnings – the NTC sample, A01, in this experiment had no reads at all which aligned to our probe kit so aligning will fail and generate warnings. This is often the case for the no template control as no biological sample should be included in that sample well. If using the GeoMX NGS Pipeline (DND) GUI a warning message should appear, or else the summary.txt file should have errors listed instead of counts for alignment of the NTC sample.
In testing, these samples took 90 minutes to process on locally installed version of the GeoMx NGS Pipeline using a system with a 3.4GHz processor, solid state hard drive and 16GB DDR2 RAM running Windows 10. NanoString recommends runs with greater than 12 FASTQ files be processed on a server installation of the Pipeline.
Glossary of terms
Raw reads: The number of reads that pass filter from the flow cell represented in the FASTQ file. GeoMx NGS Pipeline starts with FASTQ files.
Trimmed reads: Remove the adapter sequence
Stitched reads: Create one consensus read from the overlapping sequence of read 1 and read 2
Aligned reads: mapping the RTS-ID to a white list of sequences which represent targets
Deduplicated reads: remove PCR duplicates based on the UMI for digital counting
UMI: Unique Molecular Index. The molecular barcode that uniquely identifies each DNA molecule. During PCR amplification all duplicate molecules contain that barcode.
GeoMx V2.0 Software
Please make sure you are on 1.5 prior to upgrading to 2.0. If you are not on 1.5, please contact GeoMxSupport@nanostring.com.
- GeoMx DSP v2.0 Software Update
- GeoMx DSP v2.0 Update Instructions
- GeoMx DSP v2.0 Release Notes
- GeoMx DSP v2.0 Software Tips and Tricks
- Please review the GeoMx DSP v2.0 Update Instructions prior to updating the software. We recommend printing the instructions out to have it available as you perform the update.
- Please have a USB of at least 16GB and an additional laptop or computer (separate from the instrument) available to perform the update.
- Please follow the GeoMx DSP v2.0 Update Instructions to perform the update.
GeoMx NGS Pipeline v1.0
- Install Guide – Mac
- Install Guide – Windows
- Install Guide – AWS
- Install Guide – Linux
- V1.0 Release Notes
GeoMx DSPDA Off-Instrument Migration Software
For additional support or to report an issue with the software, email GeoMxSupport@nanostring.com.