SAMChain

This repository contains files to (1) push genomic data from a BAM file to a multichain blockchain, and (2)analyze data stored in the chain.

dependencies

To use this code you will need the following:

python (https://www.python.org/downloads/)

multichain (https://www.multichain.com/download-community/)

pysam (https://pysam.readthedocs.io/en/latest/installation.html)

pandas (https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html)

To develop and test this code, Python 2.7.16 and MultiChain 2.0.3 Daemon (Community Edition, latest protocol 20011) were used on Ubuntu 19.04 (GNU/Linux 5.0.0-37-generic x86_64).

buildChain.py

Use this file to initialize the chain and initialize streams within the chain One would do all of this on the command line through the following line:

python buildChain.py -cn <my-chain-name> -dr <datadir>

This will create a multichain with name in the directory .

Many other arguments can be passed to buildChain.py, including chromosome length and number, stream length, and read length. These are set to default values at the end of the file, and can be altered according to the needs of the user.

After building a chain, multichain keeps a directory <datadir>/<my-chain-name> to store the data from the chain. It also continuously runs a process for the chain.

Given a chain , one can query a position in the genome, perform depth analysis of a position, or perform pileup analysis of a position. One can also rebuild a BAM file from the read data in the chain.

insertData.py

Use this file to push data from an input BAM file to the chain. One would do all of this on the command line through the following line:

python insertData.py <mydata>.bam -cn <my-chain-name> -dr <datadir>

This will insert BAM data to a subscribed multichain with name in the directory .

queryReads.py

Use this file to query a position in the genome via the following line:

python queryReads.py <my-chain-name> chr<x>:<pos1>-<pos2> -dir <datadir> -ref <ref-genome-filename>

This will return a list of all reads in BAM format in the chain that fall in the input position range.

Example: python queryReads.py chain1 chr1:1880913 -dir ./myChains -ref GRCh38_no_alt_analysis_set_GCA_000001405.15.fa

might return this read

FC30V1RHM_20081223:8:80:961:561 0 chr1 1880913 14M1D14M CAGGCTACGAAGACAGAGTGGGGTAAAA <OPTIONAL TAGS>

queryDepth.py

Use this file to query a position in the genome for depth analysis via the following line:

python queryDepth.py <my-chain-name> chr<x>:<pos1>-<pos2> -dir <datadir> -ref <ref-genome-filename>

This will return a list of the depth statistics for all reads in the chain that fall in the input position range.

Example:

python queryDepth.py chain1 chr1:1880913-1880917 -dir ./myChains -ref GRCh38_no_alt_analysis_set_GCA_000001405.15.fa

might return these depth statistics:

chr1 1880913 1
chr1 1880914 1
chr1 1880915 1
chr1 1880916 1
chr1 1880917 1 

pileup.py

Use this file to query a position in the genome for pileup analysis via the following line:

python pileup.py <my-chain-name> chr<x>:<pos1>-<pos2> -dir <datadir> -ref <ref-genome-filename>

This will return a list of the pileup statistics for all reads in the chain that fall in the input position range.

Example:

python pileup.py chain1 chr1:1880913-1880917 -dir ./myChains -ref GRCh38_no_alt_analysis_set_GCA_000001405.15.fa

might return these pileup statistics:

chr1 1880913 C
chr1 1880914 A
chr1 1880915 G
chr1 1880916 G
chr1 1880917 C 

buildBAM.py

Use this file to build a BAM file from the read data stored in an input chain via the following line:

python buildBAM.py <my-chain-name> -dr <datadir> -bam <filename>.bam -ref <ref-genome-filename>

This will make a BAM file .bam in directory containing all read data stored in chain .

SAMchain smart contract

A sample smart contract to demonstrate one possible implementation of SAMchain in Ethereum.

---

---

INSTRUCTIONS TO RUN SAMchain SMART CONTRACT IN TRUFFLE

---

---

---

Part 1: Configuring Truffle Directories

---

1. Install Truffle via the following terminal command:

   npm install -g truffle

2. Make a directory named "truffle" and navigate to it.

3. Initialize Truffle via the following terminal command:

   truffle init

4. This will create four sub-directories and a config file: /build, /contracts, /migrations, /test, and truffle-config.js.

5. Move the smart contract solidity files to the /contracts dir.

6. Create a file in the /migrations dir named 2_deploy_contracts.js with the following contents:

   var SC = artifacts.require("./SAMchain.sol"); module.exports = function(deployer) { deployer.deploy(SC); };

---

Part 2: Running Contracts in Truffle Console

---

1. Open terminal, cd to truffle dir, run the following command:

   truffle console

2. In the truffle console, run

   truffle migrate

3. In the truffle console, run let instance = await SC.deployed()

4. Call contract methods using instance.call.

For more information, consult the Truffle Docs: https://www.trufflesuite.com/docs/truffle/getting-started/running-migrations

VCFchain

Based on SAMchain, but stores VCF rather than SAM data.

dependencies

To use this code you will need the following:

python (https://www.python.org/downloads/)

multichain (https://www.multichain.com/download-community/)

pysam (https://pysam.readthedocs.io/en/latest/installation.html)

pandas (https://pandas.pydata.org/pandas-docs/stable/getting_started/install.html)

To develop and test this code, Python 2.7.16 and MultiChain 2.0.3 Daemon (Community Edition, latest protocol 20011) were used on Ubuntu 19.04 (GNU/Linux 5.0.0-37-generic x86_64).

buildChain.py

Use this file to initialize the chain, initialize streams within the chain, and insert the VCF data. One would do all of this on the command line through the following line:

python buildChain.py -cn <my-chain-name> -dr <datadir>

This will create a multichain with name in the directory .

Many other arguments can be passed to buildChain.py. These are set to default values at the end of the file, and can be altered according to the needs of the user.

After building a chain, multichain keeps a directory <datadir>/<my-chain-name> to store the data from the chain. It also continuously runs a process for the chain.

Given a chain , one can query a position in the genome, and filter by genotype and/or variant ID.

queryAND.py

Use this file to query a variant in the genome via the following line:

python queryAND.py <my-chain-name> chr<x>:<pos1>-<pos2> -dir <datadir>

This will return a list of all variants in VCF format in the chain that fall in the input position range.