In this tutorial you will:

• Download and install spaceranger
• Key components of spaceranger folder
• Add spaceranger to $PATH
• Perform and review sitecheck
• Perform a testrun

For successful run of this tutorial, you must:

• Be comfortable in the Linux environment
• Have familiarity running command line tools
• Choose a compute platform
• Have access to a system that meets the minimum system requirements

You can download and install spaceranger in any location. For this tutorial, we will create a working directory spaceranger_tutorial and continue all the remaining steps in it.

Copy
# Create working directory
mkdir spaceranger_tutorial

# Change directory
cd spaceranger_tutorial

To install the latest version of spaceranger

• Go to the Downloads page
• Fill out the 10x Genomics End User Software License Agreement information
• Copy and paste the download command from either one of the command line utilities (curl or wget). You should see a download progress status similar to the output below.

Copy
% Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                               Dload  Upload   Total   Spent    Left  Speed
10 1220M   10  125M    0     0  30.9M      0  0:00:39  0:00:04  0:00:35 30.8M

This downloads the spaceranger tarball spaceranger-2.0.0.tar.gz to your working directory. Next, we extract the contents.

Copy
# Extract spaceranger tarball
tar -zxvf spaceranger-2.0.0.tar.gz

# Expected output
spaceranger-2.0.0/
spaceranger-2.0.0/.env.json
spaceranger-2.0.0/.version
spaceranger-2.0.0/LICENSE
spaceranger-2.0.0/builtwith.json
spaceranger-2.0.0/sourceme.bash
spaceranger-2.0.0/sourceme.csh
spaceranger-2.0.0/bin/
spaceranger-2.0.0/bin/_spaceranger_internal
spaceranger-2.0.0/bin/spaceranger
spaceranger-2.0.0/bin/rna/
spaceranger-2.0.0/bin/rna/_includes
...

When the extraction process is finished, you will have access to the command prompt and the folder spaceranger-2.0.0 will be created in the working directory.

The spaceranger-2.0.0 folder contains the executable and all of the required dependencies. The key folders that you would use are highlighted.

Copy
1spaceranger-2.0.0
2├── bin
3├── external
4│   ├── anaconda
5│   ├── martian
6│   │     └── jobmanagers
7│   ├── spaceranger_tiny_inputs
8│   └── spaceranger_tiny_ref
9├── lib
10│   ├── bin
11│   │     ├── bamtofastq
12│   │     ├── redstone
13│   │     └── ...
14│   └── python
15│       └── cellranger
16│           └── barcodes
17│               ├── visium-v1_coordinates.txt
18│               ├── visium-v2_coordinates.txt
19│               ├── visium-v4_coordinates.txt
20│               ├── visium-v5_coordinates.txt
21│               └── ...
22├── mro
23├── probe_sets
24│   ├── Visium_Human_Transcriptome_Probe_Set_v1.0_GRCh38-2020-A.csv
25│   ├── Visium_Human_Transcriptome_Probe_Set_v2.0_GRCh38-2020-A.csv
26│   └── Visium_Mouse_Transcriptome_Probe_Set_v1.0_mm10-2020-A.csv
27├── target_panels
28│   ├── gene_signature_v1.0_GRCh38-2020-A.target_panel.csv
29│   ├── immunology_v1.0_GRCh38-2020-A.target_panel.csv
30│   ├── neuroscience_v1.0_GRCh38-2020-A.target_panel.csv
31│   └── pan_cancer_v1.0_GRCh38-2020-A.target_panel.csv
32└── THIRD-PARTY-LICENSES.spaceranger.txt

• antibody_refs folder contains the validated antibody panel used in combined GEX + PEX analysis
• probe_sets contains the probe set reference CSV file used in analysis for FFPE samples
• lib/python/cellranger/barcodes folder contains visium barcodes whitelist and their coordinates on the slide
• lib/bin folder contains tools such as bamtofastq which is to convert 10x Genomics BAM files to FASTQ and redstone to enable data transfer to 10x Genomics
• external/spaceranger_tiny_ref and external/spaceranger_tiny_inputs are utilized for spaceranger testrun
• external/martian/jobmanagers folder contains sample templates for commonly used job schedulers
• THIRD-PARTY-LICENSES.spaceranger.txt file contains all the licenses for dependencies used in Space Ranger

spaceranger is now installed. There are two ways to specify spaceranger in the commands.

• Use the full path to the spaceranger-2.0.0 folder

Copy
# Method 1
## Change directory to spaceranger-2.0.0
cd spaceranger-2.0.0

## Get the full path
pwd

## Change working directory back to spaceranger_tutorial
cd ..


# Method 2
## Get the full path
readlink -f spaceranger-2.0.0


# Expected output
## The path will change dpending on the compute setup you are using.
/PATH/TO/WORKING_DIRECTORY/spaceranger_tutorial/spaceranger-2.0.0

• Adding spaceranger-2.0.0 to your $PATH variable

Copy
# Method 1
## Get the full path
readlink -f spaceranger-2.0.0

## Export PATH by providing the full path
export PATH=/PATH/TO/WORKING_DIRECTORY/spaceranger_tutorial/spaceranger-2.0.0:$PATH

## Confirm installation
which spaceranger


# Method 2
## Change directory to spaceranger-2.0.0
cd spaceranger-2.0.0

## Export PATH by specifying a shell variable
export PATH=$PWD:$PATH


## Confirm installation
which spaceranger

## Change working directory back to spaceranger_tutorial
cd ..


# Expected output
~/spaceranger_tutorial/spaceranger-2.0.0

The tilde ~ symbolizes your home directory which will be same as /PATH/TO/WORKING_DIRECTORY as before.

You can now invoke spaceranger at the command prompt to see the usage statement.

Copy
# Input
## When using full path to the spaceranger folder
/PATH/TO/WORKING_DIRECTORY/spaceranger_tutorial/spaceranger-2.0.0/spaceranger

## When adding spaceranger folder to the $PATH variable
spaceranger


# Expected output

USAGE:
spaceranger <SUBCOMMAND>

FLAGS:
    -h, --help       Prints help information
    -V, --version    Prints version information

SUBCOMMANDS:
count               Count gene expression and feature barcoding reads
                    from a single capture area
aggr                Aggregate data from multiple 'spaceranger count' runs
...
testrun             Execute the 'count' pipeline on a small test dataset
upload              Upload analysis logs to 10x Genomics support
sitecheck           Collect linux system configuration information
help                Prints this message or the help of the given subcommand(s)

For the rest of the tutorial, we will invoke spaceranger assuming addition of the spaceranger-2.0.0 folder to the $PATH variable.

spaceranger sitecheck enables you to check your system configuration to ensure it meets the minimum recommended requirements. Run the command and use > to re-direct the output to a text file.

spaceranger sitecheck > sitecheck.txt

Open the file with less and use / (e.g. /CPU Cores) to search for specific sections with the file. Press 'q' to quit.

less sitecheck.txt

We will examine some key configuration metrics and compare against the recommended system requirements.

• CPU Cores

Copy
CPU Cores
grep -c processor /proc/cpuinfo
---------------------------------------------------------------------
96
=====================================================================
...

This system has 96 CPUs and is capable of running spaceranger which requires at least 8 CPUs, preferably 32.

• Memory Total

Copy
Memory Total
grep MemTotal /proc/meminfo | cut -d ':' -f 2 | sed 's/^[ \t]*//'
---------------------------------------------------------------------
289287896 kB
=====================================================================
...

For direct comparison, let's convert kB to GB:

${\text {RAM in GB}} = \frac{289287896}{1\mathrm{e}{+6}} \approx 289$

which satisfies the system requirements of having at least [64GB RAM, preferably 128].

• User Limits

Copy
1User Limits
2bash -c 'ulimit -a'
3---------------------------------------------------------------------
4core file size          (blocks, -c) 0
5data seg size           (kbytes, -d) unlimited
6scheduling priority             (-e) 0
7file size               (blocks, -f) unlimited
8pending signals                 (-i) 1520514
9max locked memory       (kbytes, -l) 64
10max memory size         (kbytes, -m) unlimited
11open files                      (-n) 10240
12pipe size            (512 bytes, -p) 8
13POSIX message queues     (bytes, -q) 819200
14real-time priority              (-r) 0
15stack size              (kbytes, -s) 8192
16cpu time               (seconds, -t) unlimited
17max user processes              (-u) 131072
18virtual memory          (kbytes, -v) unlimited
19file locks                      (-x) unlimited
20=====================================================================

The two metrics to consider are highlighted.

a. For the max user processes, the recommendation is the limit to be 64 per core. Assuming we use all 96 cores, $96*64 = 6,144 < 131,072$

b. For max open files, the system limit $10,240 < 16,000$ which is the recommendation. While the pipelines may run at lower open file limit, caution is urged. This value is dependent on the system, the sample type and number of samples being run. In case the pipeline errors, it is advisable to increase the user limit ulimit and try again.

• Global File Limit

Copy
Global File Limit
cat /proc/sys/fs/file-{max,nr}
---------------------------------------------------------------------
2921445
68736   0       262144
=====================================================================

The value satisfies the minimum requirement of 10k per GB RAM $10,000*289 = 2,890,000 < 2,921,445$, where 289 GB is the total memory of the system. The software support team can review your sitecheck results. There are two ways to send it across:

• If the compute platform has access to internet, use the upload pipeline replacing the email address with your email spaceranger upload [email protected] sitecheck.txt
• If the compute platform is not connected to the internet, you can send the sitecheck.txt as an attachment to [email protected].

We can verify the installation using spaceranger testrun. This pipeline can be run in two configurations depending on the internet connectivity of the compute platform.

Copy
# With internet access
spaceranger testrun --id=verify_install

# Without internet access
spaceranger testrun --no-internet --id=verify_install

# Expected output
Martian Runtime - v4.0.5

Running preflight checks (please wait)...
Checking sample info...
Checking FASTQ folder...
Checking reference...
Checking reference_path
Checking optional arguments...

...

Pipestance completed successfully!

Successful completion of the testrun by extension implies successful installation of spaceranger.

Q: How can I use multiple versions of spaceranger?
Sometimes it is useful to have access to older as well as newer versions of spaceranger. There are two suggested ways to achieve this:

• Update $PATH to point to the latest version

Since the spaceranger tarball comes annotated with the version number, you can download and uninstall the latest version and subsequently update the [$PATH] variable to point to the version you wish to use.

• Use virtual environments

You can install and set up conda which functions as both package and environment manager. Use of virtual environments for running software provides many useful benefits such as reproducibility, compatibility, versioning as well as giving admin permissions on shared compute environments such as High-Performance Computing clusters (HPCs).