Overview

Product video
What it does DISTILL covers the whole omics research loop in one place. It indexes the public record, starting with hundreds of thousands of GEO studies and adding PRIDE proteomics and more repositories, into a fast, locally hosted catalogue. Researchers can go from which datasets profile CD8+ T cells in heart failure? to a processed cohort in minutes instead of weeks. Studies from public repositories or your own labs run through the same reproducible pipelines for quality control, clustering, cell-type annotation, differential expression and batch-corrected multi-study integration. The pipelines cover single-cell RNA-seq, spatial transcriptomics, and mass-spectrometry and affinity proteomics. On top of this, AI agents search, process, compare and test hypotheses across studies and data types. Every result becomes a shareable, traceable report rather than a chat answer that disappears. Each study you process also feeds a disease-centric knowledge base linking genes, cell types and conditions, so DISTILL becomes your organizations growing memory of its biology rather than a study-by-study viewer.
Why teams choose DISTILL Multi-modal by design Bring together single-cell and bulk RNA-seq, spatial omics, proteomics, epigenomics, Perturb-seq, and clinical data to build a more complete view of biology. Agentic analysis, built around the scientific question Specialized AI agents coordinate analysis, interpretation, and exploration across datasets, adapting workflows as research questions evolve. Explore, iterate, and collaborate Revisit analyses, test new hypotheses, compare alternative interpretations, and share observations with colleagues across biology, data science, and clinical teams. Connect insights across studies and discover recurring signals Compare findings across studies, cohorts, conditions, treatments, and platforms to identify recurring biological signals, connect evidence, and generate stronger hypotheses. From analysis to scientific communication Turn analyses and observations into structured, publication-ready reports with the supporting context and evidence captured along the way.
Highlights
- AI-powered analytics platform that streamlines scRNA-seq interpretation for faster mechanism and biomarker discovery.
Details
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Cost/unit |
|---|---|
study | $5,000.00 |
Vendor refund policy
For refund requests, contact finance@aganitha.ai
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
DISTILL on Amazon EKS
- Amazon EKS
Container image
Containers are lightweight, portable execution environments that wrap server application software in a filesystem that includes everything it needs to run. Container applications run on supported container runtimes and orchestration services, such as Amazon Elastic Container Service (Amazon ECS) or Amazon Elastic Kubernetes Service (Amazon EKS). Both eliminate the need for you to install and operate your own container orchestration software by managing and scheduling containers on a scalable cluster of virtual machines.
Version release notes
DISTILL 1.1.1.
No container runs as root by default: Redis and the Traefik ingress now run as non-root users, and every image the product runs is delivered from AWS Marketplace.
Also includes everything in 1.1.0.
Before you deploy, you will need: a domain with DNS you control, a wildcard TLS certificate for it, an SMTP account, an NCBI API key, and a Google Gemini API key. See the usage instructions for details.
Additional details
Usage instructions
Usage Instructions All DISTILL containers run as non-root users by default and do not require privileged mode or root access.
Step 1: AWS permissions required
The AWS user deploying this software must be able to create every resource the stack provisions. We recommend using a least-privilege IAM policy that allows only the following services and actions:
cloudformation, ec2 (VPC, subnets, security groups, instances), eks, rds, elasticfilesystem, fsx, lambda, route53, secretsmanager, s3, logs, cloudwatch, sqs, sns, dynamodb, and license-manager / aws-marketplace metering actions, plus iam for roles, policies, instance profiles and OIDC providers.
If permissions are missing the stack fails during creation and the CloudFormation console names the action that was denied. Grant it, delete the stack, and retry.
Step 2: EC2 service quotas (request before deploying)
AWS limits how many vCPUs your account may run at once. A new account allows 5 - fewer than this stack needs to install. Raise both limits below under Service Quotas > Amazon EC2 in the region you will deploy to. Increases are free; you pay only for what runs.
Running On-Demand Standard (A, C, D, H, I, M, R, T, Z) instances 10 vCPU to install, 170 to run the compute cluster at full scale. Below 10 the deployment fails while creating the EKS node.
Running On-Demand G and VT instances 80 vCPU. Every account starts at 0, so without this GPU pipeline steps cannot run. AWS reviews GPU requests manually - allow several days.
Figures assume the default instance types; smaller ones lower them.
Step 3: Our quick create link, backed by our CloudFormation template, has several parameters you must supply values for. Some of them require preparation first:
a. NcbiEmail and NcbiApiKey: Follow the instructions on NCBI's official portal to generate an API key - https://ncbiinsights.ncbi.nlm.nih.gov/2017/11/02/new-api-keys-for-the-e-utilities/
b. SmtpEmailAddress and SmtpPassword: Set up your SMTP server and supply its credentials in these parameters. The application uses them for its authentication system.
c. GoogleApiKey: Generate a Gemini API key from Google and supply it here. This enables the application's LLM capabilities.
d. GlobalDomain: The domain you want to host the application on. Make sure the DNS server for that domain is active.
e. The application consists of several microservices, each needing a subdomain name: ApiHostName, JupyHostName, AgentHostName and FrontendHostName.
f. TraefikCertPem and TraefikKeyPem: Generate a wildcard SSL certificate and key for your domain, then supply each one base64-encoded rather than as raw PEM:
base64 -i cert.pem | tr -d '\n' base64 -i key.pem | tr -d '\n'
Paste each resulting single line into the matching parameter. Base64 because a CloudFormation parameter field is single-line: a raw PEM pasted into it loses its newlines and becomes unreadable. This gives the hosted services HTTPS.
The remaining parameters need no external setup.
Step 4: Once you have those values, open our quick create link from the deployment templates section and deploy the stack. You can follow its progress in the CloudFormation console. Stack creation fails if the deployer lacks the permissions described in Step 1; the CloudFormation console will say so. Grant the missing permission, delete the stack, and create it again.
Step 5: Post-deployment: update your DNS records:
Once the stack is deployed, open the Outputs tab of the main (parent) CloudFormation stack. You will find the Traefik load balancer URL there. In your DNS server, point the hostnames you supplied as stack parameters at that load balancer URL. You can then log in at https://{FrontendHostName}.{GlobalDomain}: for example frontend.example.com, if you set the frontend hostname to 'frontend' and your global domain to 'example.com'.
Resources
Vendor resources
Support
Vendor support
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.