All new
Data Science
jobs, in one place.

Updated daily to help you be the first to apply ⏱

Postdoc Fellow – Developing machine learning predictive models using chromatographic data
  • Python
  • Java
  • Machine Learning
  • Database
  • Deep Learning
  • PyTorch
Cambridge CB4
107 days ago

AstraZeneca’s drug discovery teams design and synthesise thousands of novel compounds every year. As part of this process, chromatographic and mass spectrometric data is collected for each molecule resulting in large datasets of experimental data. As part of our efforts to learn from this in-house experimental data, AstraZeneca has developed internal analytical databases, containing details of chromatographic separation conditions for many thousands of diverse molecules. These include data gathered using reverse-phase HPLC and both achiral and chiral SFC experiments.

This postdoc provides the unique opportunity to apply state-of-the-art machine learning techniques to mine this high-quality dataset for new knowledge to impact future molecule analysis and purification on real drug discovery projects. The models will be trained on relevant computational molecular descriptors for the compounds, mobile and stationary phases in

chromatographic systems and can also incorporate other experimentally derived physicochemical parameters (e.g. logD, pKa, ePSA) with the aim to predict chromatographic behaviour. Key endpoints will include prediction of retention time and mobile and stationary phase conditions required for optimised resolution of reaction products. Through model interpretation, this project will also aim understand the mechanism and strength of binding interactions between specific functional groups of both analyte and stationary phase within chromatographic systems.

This project will be supervised by Jennifer Kingston within the AZ separation sciences groups in Cambridge, as well as by Prof. Jonathan Goodman, a leading academic in the fields of cheminformatics and machine learning from the University of Cambridge, allowing the successful candidate to benefit from both academic and industrial environments. This exciting opportunity will involve continuous exchange between modelling and experimental teams to enable the experimental validation of models as well as to deliver impact on real drug projects. We plan to publish the results of the study, contributing to this rapidly growing field.

Do you want to be part of this exciting project? If so, don't hesitate in applying today!!

Education and Experience required:


  • PhD in computational chemistry, cheminformatics or machine learning.
  • Knowledge of programming (e.g. Python, C++, Java).
  • Experience in applying and validating machine learning algorithms on real datasets.
  • A proven record of productivity and problem-solving ability.
  • Interest in experimental methods for chemical characterisation.


  • Fluency in a deep learning framework such as Tensorflow or PyTorch.
  • Experience in calculating descriptors for encoding the 2D and 3D properties of molecules.
  • Background in Chemistry or knowledge of chromatographic techniques, compound purification or synthetic chemistry.
  • Knowledge of molecular modelling software packages.
  • Strong publication track record.

Competitive flexible benefits and generous remuneration apply!

    Related Jobs

  • Data Scientist

    • Python
    • SQL
    • Scala
    Lloyds Banking Group
  • Graduate Data Scientist - immediate start

    • Python
    • Machine Learning
  • Data Scientist - Economist

    • Python
    • SQL
    • Machine Learning
    7 days ago
  • Assistant Data Analyst

    • Business Intelligence
    The University of Manchester
    Manchester M1
    1 day ago
  • Senior Data Analyst

    • Tableau
    • Database
    1 day ago