Create large-scale ML-driven multiscale simulation ensembles to study the interactions

Overview

MuMMI RAS v0.1

Released: Nov 16, 2021

MuMMI RAS is the application component of the MuMMI framework developed to create large-scale ML-driven multiscale simulation ensembles to study the interactions of RAS proteins and RAS-RAF protein complexes with lipid plasma membranes.

MuMMI framework was developed as part of the Pilot2 project of the Joint Design of Advanced Computing Solutions for Cancer funded jointly by the Department of Energy (DOE) and the National Cancer Institute (NCI).

The Pilot 2 project focuses on developing multiscale simulation models for understanding the interactions of the lipid plasma membrane with the RAS and RAF proteins. The broad computational tool development aims of this pilot are:

  • Developing scalable multi-scale molecular dynamics code that will automatically switch between phase field, coarse-grained and all-atom simulations.
  • Developing scalable machine learning and predictive models of molecular simulations to:
    • identify and quantify states from simulations
    • identify events from simulations that can automatically signal change of resolution between phase field, coarse-grained and all-atom simulations
    • aggregate information from the multi-resolution simulations to efficiently feedback to/from machine learning tools
  • Integrate sparse information from experiments with simulation data

MuMMI RAS defines the specific functionalities needed for the various components and scales of a target multiscale simulation. The application components need to define the scales, how to read the corresponding data, how to perform ML-based selection, how to run the simulations, how to perform analysis, and how to perform feedback. This code uses several utilities made available through "MuMMI Core".

Publications

MuMMI framework is described in the following publications.

  1. Bhatia et al. Generalizable Coordination of Large Multiscale Ensembles: Challenges and Learnings at Scale. In Proceedings of the ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC '21, Article No. 10, November 2021. doi:10.1145/3458817.3476210.

  2. Di Natale et al. A Massively Parallel Infrastructure for Adaptive Multiscale Simulations: Modeling RAS Initiation Pathway for Cancer. In Proceedings of the ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC '19, Article No. 57, November 2019. doi:10.1145/3295500.3356197.
    Best Paper at SC 2019.

  3. Ingólfsson et al. Machine Learning-driven Multiscale Modeling Reveals Lipid-Dependent Dynamics of RAS Signaling Protein. Proceedings of the National Academy of Sciences (PNAS), accepted, 2021. preprint.

  4. Reciprocal Coupling of Coarse-Grained and All-Atom scales. In preparation.

Installation

git clone https://github.com/mummi-framework/mummi-ras
cd mummi-ras
pip3 install .

export MUMMI_ROOT=/path/to/outputs
export MUMMI_CORE=/path/to/core/repo
export MUMMI_APP=/path/to/app/repo
export MUMMI_RESOURCES=/path/to/resources
The installaton process as described above installs the MuMMI framework. The simulation codes (gridsim2d, ddcMD, AMBER, GROMACS) are not included and are to be installed separately.
Spack installation. We are also working towards releasing the option of installing MuMMI and its dependencies through spack.

Authors and Acknowledgements

MuMMI was developed at Lawrence Livermore National Laboratory, in collaboration with Los Alamos National Laboratory, Oak Ridge National Laboratory, and International Business Machines. A list of main contributors is given below.

  • LLNL: Harsh Bhatia, Francesco Di Natale, Helgi I Ingólfsson, Joseph Y Moon, Xiaohua Zhang, Joseph R Chavez, Fikret Aydin, Tomas Oppelstrup, Timothy S Carpenter, Shiv Sundaram (previously LLNL), Gautham Dharuman (previously LLNL), Dong H Ahn, Stephen Herbein, Tom Scogland, Peer-Timo Bremer, and James N Glosli.

  • LANL: Chris Neale and Cesar Lopez

  • ORNL: Chris Stanley

  • IBM: Sara K Schumacher

MuMMI was funded by the Pilot2 project led by Dr. Fred Streitz (DOE) and Dr. Dwight Nissley (NIH). We acknowledge contributions from the entire Pilot 2 team.

This work was performed under the auspices of the U.S. Department of Energy (DOE) by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344, Los Alamos National Laboratory (LANL) under Contract DE-AC5206NA25396, and Oak Ridge National Laboratory under Contract DE-AC05-00OR22725.

Contact: Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550.

Contributing

Contributions may be made through pull requests and/or issues on github.

License

MuMMI RAS is distributed under the terms of the MIT License.

Livermore Release Number: LLNL-CODE-827655

Comments
  • Are the trajectories in your publications publicly available?

    Are the trajectories in your publications publicly available?

    Hi, Congrats on the success, and huge thanks for making it open source. I wonder whether the trajectories in your publications are publicly available. Or are there any demo trajectories?

    I am a Ph.D. student at KAUST, using computer graphics to build and visualize mesoscale biology models, such as SARS-CoV-2 and bacteriophage T4. If possible, I (and my colleagues) would like to perform (multiscale, multi-representation, multi-granularity) visualization research on the trajectories you generated.

    Many thanks, Roden

    opened by RodenLuo 2
  • `flux` vs `slurm`

    `flux` vs `slurm`

    Hi,

    As flux is mentioned in the dependencies, is it possible to reproduce MuMMI RAS on a cluster that only has slurm?

    Workflow dependencies (e.g., python, flux, dynim, keras, etc.)

    Quoted from: https://github.com/mummi-framework/mummi-ras/blob/main/INSTALL.md

    Many thanks, Roden

    opened by RodenLuo 0
  • gridsim2d availability

    gridsim2d availability

    Hi, I wonder if the following code is available or not.

    gridsim2d: to be released shortly

    Quoted from: https://github.com/mummi-framework/mummi-ras/blob/main/INSTALL.md

    Thanks, Roden

    opened by RodenLuo 0
  • Patch for gromacs availability

    Patch for gromacs availability

    Hi, I wonder if the following patch is available or not.

    Note that we have a patch for gromacs installation for customization. To be open-sourced soon.

    Quoted from: https://github.com/mummi-framework/mummi-ras/blob/main/INSTALL.md

    Thanks, Roden

    opened by RodenLuo 0
  • Small scale test data for local deployment

    Small scale test data for local deployment

    Hi, I'm interested in deploying MuMMI on the KAUST IBEX cluster. It is mentioned in the installation doc that there is a small set of test data. Is it now publicly available? If not, is it possible for me to somehow access it so that I can perform a test run?

    Many thanks, Roden

    Again on lassen and on summit, we have created a small set of test data, which can be used to launch MuMMI at small scales. This (and the larger dataset) will be made public through NCI website. Until then, we can make this data available upon request.

    opened by RodenLuo 1
Releases(v1.0.0)
Fast Fourier Transform-accelerated Interpolation-based t-SNE (FIt-SNE)

FFT-accelerated Interpolation-based t-SNE (FIt-SNE) Introduction t-Stochastic Neighborhood Embedding (t-SNE) is a highly successful method for dimensi

Kluger Lab 547 Dec 21, 2022
Predicting job salaries from ads - a Kaggle competition

Predicting job salaries from ads - a Kaggle competition

Zygmunt Zając 57 Oct 23, 2020
A toolbox to iNNvestigate neural networks' predictions!

iNNvestigate neural networks! Table of contents Introduction Installation Usage and Examples More documentation Contributing Releases Introduction In

Maximilian Alber 1.1k Jan 05, 2023
In this Repo a simple Sklearn Model will be trained and pushed to MLFlow

SKlearn_to_MLFLow In this Repo a simple Sklearn Model will be trained and pushed to MLFlow Install This Repo is based on poetry python3 -m venv .venv

1 Dec 13, 2021
Predicting Baseball Metric Clusters: Clustering Application in Python Using scikit-learn

Clustering Clustering Application in Python Using scikit-learn This repository contains the prediction of baseball metric clusters using MLB Statcast

Tom Weichle 2 Apr 18, 2022
WAGMA-SGD is a decentralized asynchronous SGD for distributed deep learning training based on model averaging.

WAGMA-SGD is a decentralized asynchronous SGD based on wait-avoiding group model averaging. The synchronization is relaxed by making the collectives externally-triggerable, namely, a collective can b

Shigang Li 6 Jun 18, 2022
Production Grade Machine Learning Service

This project is made to help you scale from a basic Machine Learning project for research purposes to a production grade Machine Learning web service

Abdullah Zaiter 10 Apr 04, 2022
PySpark ML Bank Churn Prediction

PySpark-Bank-Churn Surname: corresponds to the record (row) number and has no effect on the output. CreditScore: contains random values and has no eff

kemalgunay 2 Nov 11, 2021
A Python package for time series classification

pyts: a Python package for time series classification pyts is a Python package for time series classification. It aims to make time series classificat

Johann Faouzi 1.4k Jan 01, 2023
Machine Learning Course with Python:

A Machine Learning Course with Python Table of Contents Download Free Deep Learning Resource Guide Slack Group Introduction Motivation Machine Learnin

Instill AI 6.9k Jan 03, 2023
Programming assignments and quizzes from all courses within the Machine Learning Engineering for Production (MLOps) specialization offered by deeplearning.ai

Machine Learning Engineering for Production (MLOps) Specialization on Coursera (offered by deeplearning.ai) Programming assignments from all courses i

Aman Chadha 173 Jan 05, 2023
Code Repository for Machine Learning with PyTorch and Scikit-Learn

Code Repository for Machine Learning with PyTorch and Scikit-Learn

Sebastian Raschka 1.4k Jan 03, 2023
A Pythonic framework for threat modeling

pytm: A Pythonic framework for threat modeling Introduction Traditional threat modeling too often comes late to the party, or sometimes not at all. In

Izar Tarandach 644 Dec 20, 2022
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.

DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective. 10x Larger Models 10x Faster Trainin

Microsoft 8.4k Dec 30, 2022
Simple, light-weight config handling through python data classes with to/from JSON serialization/deserialization.

Simple but maybe too simple config management through python data classes. We use it for machine learning.

Eren Gölge 67 Nov 29, 2022
Provide an input CSV and a target field to predict, generate a model + code to run it.

automl-gs Give an input CSV file and a target field you want to predict to automl-gs, and get a trained high-performing machine learning or deep learn

Max Woolf 1.8k Jan 04, 2023
This project used bitcoin, S&P500, and gold to construct an investment portfolio that aimed to minimize risk by minimizing variance.

minvar_invest_portfolio This project used bitcoin, S&P500, and gold to construct an investment portfolio that aimed to minimize risk by minimizing var

1 Jan 06, 2022
YouTube Spam Detection with python

YouTube Spam Detection This code deletes spam comment on youtube videos based on two characteristics (currently) If the author of the comment has a se

MohamadReza Taalebi 5 Sep 27, 2022
fMRIprep Pipeline To Machine Learning

fMRIprep Pipeline To Machine Learning(Demo) 所有配置均在config.py文件下定义 前置环境(lilab) 各个节点均安装docker,并有fmripre的镜像 可以使用conda中的base环境(相应的第三份包之后更新) 1. fmriprep scr

Alien 3 Mar 08, 2022