MolSSI Workshop: Machine Learning and Chemistry: Challenges on the Way Forward

Organizers: Pratyush Tiwary (U. Maryland), Olexandr Isayev (Carnegie Mellon U.), Adrian Roitberg (U. Florida)

Location:  College Park, Maryland

Dates: 16-18 November 2019

In recent years, the field of machine learning (ML) has seen an incredible surge in interest. From image classifiers to board games, ML and big data and internet are making large impacts in nearly every field. Chemistry and molecular science in general are no different. ML and informatics techniques have demonstrated strong success in different aspects of chemistry. The success stories so far include, but are not limited to (a) facilitating extremely high throughput screening methods for understanding complex chemical processes and for formulating materials with desirable properties, (b) automated enhanced sampling approaches for tackling the rare event problem in classical and ab initio simulations, (c) speeding up the rational design of inhibitors for treating complex diseases. These are just a few examples, and the number of successful instances where ML based approaches have started to change the status quo in chemistry is increasing very rapidly. However, despite the rate of advancement in this field, the groundbreaking potential of these approaches is yet to be fully realized. There are few or no conferences or workshops dedicated to the application of these methods to chemical and material problems, where practitioners in different sub-disciplines of chemistry can get together and learn from each other.

In order to stimulate further improvements, we will hold a dual-purpose workshop. First, there will be cross-talk between experts using ML in different aspects of chemistry, such as designing models of molecular interactions, drug design, enhanced sampling, designing chemical synthesis routes and others. These will be aided by extensive discussion time after different talks as well as brainstorming sessions. Second, the workshop will have a strong emphasis on considering questions concerning the software side of using ML in chemistry. These would include for example:

  • how to implement public access to reliable benchmark data to enforce reproducibility;
  • what software tools & infrastructures are missing for chemists to use ML in everyday work?;
  • how to train the next generation of chemistry students who can use such ML software; and
  • open-source standards that need to be set and met, which can often be challenged by the intellectual property rights related to these novel and important approaches.


  • Robert Abel (Schrodinger)
  • Robert Best (NIH)
  • Michelle Ceriotti (EPFL)
  • John Chodera (MSKCC)
  • Gaurav Chopra (Purdue University)
  • Paulette Clancy (Johns Hopkins)
  • Lucy Colwell, Cambridge University
  • Amir Barati Farimani, Carnegie Mellon University
  • Rafael Gomez-Bombarelli (MIT)
  • Johannes Hachmann (SUNY Buffalo)
  • Gerhard Hummer (MPI Frankfurt)
  • Shantenu Jha (Rutgers)
  • Boris Kozinsky (Harvard)
  • Heather Kulik (MIT)
  • Tom Miller (Caltech)
  • Tim Mueller (Johns Hopkins)
  • Michele Parrinello (ETH/USI Switzerland and IIT Genova)
  • Sapna Sarupria (Clemson)
  • Ruhong Zhou (IBM)

The final program of this workshop can be found HERE.