Genomics Tertiary Analysis and Machine Learning Using Amazon SageMaker

The Genomics Tertiary Analysis and Machine Learning Using Amazon SageMaker solution creates a scalable environment in AWS to develop machine learning models using genomics data, generate predictions, and evaluate model performance. This solution demonstrates how to 1) automate the preparation of a genomics machine learning training dataset, 2) develop genomics machine learning model training and deployment pipelines and, 3) generate predictions and evaluate model performance using test data.

Standard deployment

To deploy this solution in your account use the "Launch in the AWS Console" button found on the solution landing page.

We recommend deploying the solution this way for most use cases.

Customized deployment

A fully customized solution can be deployed for the following use cases:

• Modifying or adding additional resources deployed during installation
• Modifying the "Landing Zone" of the solution - e.g. adding additional artifacts or customizing the "Pipe" CodePipeline

Fully customized solutions need to be self-hosted in your own AWS account, and you will be responsible for any costs incurred in doing so.

To deploy and self-host a fully customized solution use the instructions below.

Note: All commands assume a bash shell.

Customize

Clone the repository, and make desired changes

File Structure

.
├── CHANGELOG.md
├── CODE_OF_CONDUCT.md
├── CONTRIBUTING.md
├── LICENSE.txt
├── NOTICE.txt
├── README.md
├── buildspec.yml
├── deploy.sh
├── deployment
│   ├── build-open-source-dist.sh
│   ├── build-s3-dist.sh
│   └── run-unit-tests.sh
└── source
    ├── GenomicsLearningCode
    │   ├── awscli_test.sh
    │   ├── code_cfn.yml
    │   ├── copyresources_buildspec.yml
    │   ├── resources
    │   │   ├── notebooks
    │   │   │   ├── variant_classifier-autopilot.ipynb
    │   │   │   └── variant_predictor.ipynb
    │   │   └── scripts
    │   │       └── process_clinvar.py
    │   └── setup
    │       ├── lambda.py
    │       └── requirements.txt
    ├── GenomicsLearningPipe
    │   └── pipe_cfn.yml
    ├── GenomicsLearningZone
    │   └── zone_cfn.yml
    ├── setup.sh
    ├── setup_cfn.yml
    └── teardown.sh

Run unit tests

cd ./deployment
chmod +x ./run-unit-tests.sh
./run-unit-tests.sh

Build and deploy

Create deployment buckets

The solution requires two buckets for deployment:

1. <bucket-name> for the solution's primary CloudFormation template
2. <bucket-name>-<aws_region> for additional artifacts and assets that the solution requires - these are stored regionally to reduce latency during installation and avoid inter-regional transfer costs

Configure and build the distributable

export DIST_OUTPUT_BUCKET=<bucket-name>
export SOLUTION_NAME=<solution-name>
export VERSION=<version>

chmod +x ./build-s3-dist.sh
./build-s3-dist.sh $DIST_OUTPUT_BUCKET $SOLUTION_NAME $VERSION

Deploy the distributable

Note: you must have the AWS Command Line Interface (CLI) installed for this step. Learn more about the AWS CLI here.

cd ./deployment

# deploy global assets
# this only needs to be done once
aws s3 cp \
    ./global-s3-assets/ s3://<bucket-name>/$SOLUTION_NAME/$VERSION \
    --recursive \
    --acl bucket-owner-full-control

# deploy regional assets
# repeat this step for as many regions as needed
aws s3 cp \
    ./regional-s3-assets/ s3://<bucket-name>-<aws_region>/$SOLUTION_NAME/$VERSION \
    --recursive \
    --acl bucket-owner-full-control

Install the customized solution

The link to the primary CloudFormation template will look something like:

https://<bucket-name>.s3-<region>.amazonaws.com/genomics-tertiary-analysis-and-data-lakes-using-amazon-sagemaker.template

Use this link to install the customized solution into your AWS account in a specific region using the AWS Cloudformation Console.

This solution collects anonymous operational metrics to help AWS improve the quality of features of the solution. For more information, including how to disable this capability, please see the implementation guide.

Copyright 2019 Amazon.com, Inc. or its affiliates. All Rights Reserved.

Licensed under the Apache License Version 2.0 (the "License"). You may not use this file except in compliance with the License. A copy of the License is located at

http://www.apache.org/licenses/

or in the "license" file accompanying this file. This file is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, express or implied. See the License for the specific language governing permissions and limitations under the License.