Nextflow
Run Nextflow Pipelines at Scale
Hundreds of biotechs use Latch to make data analysis faster, cheaper, more accessible, and instantly accelerate their R&D milestones.
The best place to run Nextflow pipelines
Latch solves the issues hosting infrastructure for Nextflow pipelines and delivering workflows to scientists.
User-friendly, Type-safe Interfaces
Latch’s integration with Nextflow allows developers to build user-friendly graphical interfaces for their workflows and expose them to wet lab scientists.
No changes required to existing Nextflow code
Customizable interface via an additional Python file
Bring data from anywhere - AWS, GCP, SRA, BaseSpace, and more.
Built for Developers
Nextflow on Latch provides managed cloud infrastructure for executing, debugging, and analyzing workflows.
It uses a customized Nextflow Kubernetes plugin to run processes in containers within Latch’s Kubernetes cluster, which allows developers to use existing Nextflow projects with minimal code alterations.
Intermediate outputs are shared via AWS EFS, and final outputs can be uploaded to Latch Data using a custom plugin.
The Latch Console features a Directed Acyclic Graph (DAG) for monitoring progress, checking logs, and analyzing process costs and runtime. Developers also have real-time access to the container for debugging purposes.
Robust toolings for debugging
Latch makes it easy to trace logs and processes from which they originate, even after a pipeline finishes and its machine has been cleaned up.
Cost and Performance Reporting
No more manually calculating costs. Nextflow on Latch has built-in real time cost reports with every workflow, making profiling and pipeline optimization a breeze for developers.
Downstream analysis tooling
With Latch, scientists are no longer left with just raw output files in cloud storage systems. Easily pull results into interactive visualizations to make critical scientific decisions.
Results to Plots
Downstream analyses in Pods
Native visualizations (FastQC, CELLxGENE) for common files
How it works
Upload your own
Install Latch & Clone your Nextflow repository
$pip install latch$git clone https://github.com/nf-core/rnaseq$cd rnaseqDefine metadata and workflow graphical interface
$latch generate-metadata nextflow_schema.json --nextflowMetadata files:
latch_metadata/__init__.py
latch_metadata/parameters.py
Register the workflow to Latch
$latch login$latch register . --nf-script main.nf --nf-execution-profile docker– and that’s it. You just uploaded the nf-core/rnaseq pipeline to Latch!
Or access the nf-core library on Latch
nf-core/rnaseq
nf-core/sarek
nf-core/scrnaseq
nf-core/methylseq
nf-core/mag
nf-core/atacseq
Each pipeline on Latch has a user-friendly interface, a graphical, error-validated sample sheet component to fill out pipeline inputs, and a directed acyclic graph (DAG) to view process and sample-specific errors.
For Bulk RNA-seq, ATAC-seq, and Methyl-seq, Latch also offers interactive plotting dashboards that directly ingest NF-core outputs and produce publication-ready figures.
Features
Performance Optimization
Easily view and compare results, cost, and performance of different executions
Graphical workflow interface
Easily view and compare results, cost, and performance of different executions
Integration with Latch Registry
Developers can extend their existing Nextflow workflow to read from Latch Registry, a user-friendly, error-validated sample sheet input system.
Debugging
Avoid failed workflows caused by discrepancies between running code locally and in the cloud.
Latch SDK comes with the latch develop command, which drops you into an interactive shell where you can run your code and inspect the environment before registering and executing the entire workflow in the cloud to debug environments and logical issues.
Relaunching from failed tasks
Nextflow's integration uses caching to save intermediate workflow results, allowing for relaunch without full re-execution.
Latch also exposes storage expiration hours as a configurable parameter to users to determine how long the cache will persist after a failure.Â
Logging
To facilitate debugging, a DAG of processes is generated in the Latch console to enable developers to easily track progress, view logs, and analyze cost and runtime of each process execution.
EFS integration
Intermediate outputs are communicated between processes via AWS EFS and published outputs can be uploaded directly to Latch Data via a custom filesystem plugin. The work directory is automatically cleaned up after workflow completion, saving costs and eliminating the hassle of manually cleaning up resources
Ready-to-use NF-core pipelines
Latch has top 10 ready-to-launch NF-core workflows with user-friendly interfaces, documentation, and GitHub repositories to help you get started.
Sharing of pipelines
There’s no seat limit on Latch. Everyone with a Latch account can create any number of workspaces and add unlimited team members to each workspace. As an Admin, user can also distribute versions of their Nextflow pipelines to multiple workspaces.
Private registries
Latch Secrets allow Nextflow processes on Latch to securely use containers images hosted in users’ private registries (GitLab, AWS ECR, Azure container registry, and more)
Integration with GitHub
Workflows on Latch can be versioned using Git. If the SDK detects that the project directory is a Git repository, it will append the first six digits of the latest commit hash to the workflow version.
Pay-as-you-go. No commitment upfront.
There’s no license fee on Latch. You can add unlimited number of users on the platform, and only pay for compute and storage.
Learn More →7-Day Free Trial
Access the leading data platform for biology.
- A scoping call to map out your scientific workflows
- Access a curated workspace customized to your assays and needs
- Live 16/6 support with a Latch bioinformatics engineer
- A hands-on onboarding session to train your team