Postdoc in Machine Learning for Cancer Pharmacogenomics

APPLICATION CLOSING DATE: July 3rd, 2023

Human Technopole (HT) is a new interdisciplinary life science research institute, created and supported by the Italian Government, with the aim of developing innovative strategies to improve human health. HT is composed of five Centers: Computational Biology, Structural Biology, Genomics, Neurogenomics and Health Data Science. The Centers work together to enable interdisciplinary research and to create an open, collaborative environment that will help promote life science research both nationally and internationally.

The Iorio Lab in the Computational Biology Research Centre at Human Technopole in Milan works at the interface of biology, machine learning, statistics and information theory; our goal is to understand and predict how genomic alterations and molecular traits from other omics contribute to pathological processes, biological circuits’ rewiring and have an impact on therapeutic response in human cancers and other diseases. Our research aims at advancing human health by designing algorithms, computational tools and novel analytical methods for the integration and the analysis of pharmacogenomics and functional-genomics datasets, with the ultimate objective of identifying new therapeutic targets, biomarkers and drug repositioning opportunities.

We are looking for a highly motivated postdoctoral researcher with strong skills in Computational Biology to fill a Postdoc- position in the research team led by Francesco Iorio within the Centre for Computational Biology. This role will be actively interacting with the other HT research centres, as well as national and international collaborators, such as those involved in the Cancer Dependency Map partnership, whose broad goal is to systematically identify cancer vulnerabilities and dependencies that could be exploited therapeutically.

The selected candidate will work on the development of new algorithms, analytical methods and tools for the analysis of data from perturbational screens performed on cancer pre-clinical models (cell-lines, organoids and patient derived xerographs) with various phenotypic readout (ranging from cell viability reduction, to sc-transcriptomics and spatial transcriptomics); will develop machine learning tools for the prediction of cancer dependencies and drug response from the integration of the aforementioned screens and the multiomic characterisation of the screened models; will interact with other research groups and prepare research products for manuscript submission and presentations to international meetings/conferences.

Key tasks and responsibilities:

• Performing original computational biology research and developing novel methods and models to identify clinically relevant cancer dependencies and vulnerabilities.
• Developing and applying new tools for the integrative analysis and visualization of large multiomic / multimodal biomedical datasets,
• Designing, implementing, and explaining predictive models based on statistical inference and machine learning approaches to identify molecular determinants of gene essentiality and drug response.
• Contributing to the analysis and interpretation of multidisciplinary collaborative projects with translational and clinical collaborators.
• Contributing to study design and research project management.

Job requirements

Mandatory requirements:

• PhD in Computational Biology, Bioinformatics, Statistics, Computer Science, or related fields.
• Strong quantitative data analysis and visualization expertise.
• Ability to own planning, execution, and delivery of scientific results.
• Full proficiency in at least one scripting language (Python, R, Julia).

Preferential requirements:

• Experience in developing and implementing statistical and machine learning based predictive models.
• Familiarity with major public sources of transcriptomic and genomic data.
• Excellent verbal, written, analytical, and presentation communication skills.
• Fluent in English.

Other competencies:

• Solid understanding of molecular biology and oncology.
• Previous experience in single-cell omic data analysis, especially single-cell RNAseq.
• Familiarity with cluster computing: Slurm, Nexflow, Snakemake.
• Familiarity with graph databases and graph neural networks.
• Programming languages: full scripting proficiency with both Python and R.
• Python packages: PyTorch, Numba, CUDA.
• Familiarity with software development best practices: testing, versioning, CI/CD.
• Strong publishing record in Computational Biology, Bioinformatics journals.
• Experience in contributing to the design of collaborative multidisciplinary projects.

Application Instructions

To apply, please send the following:

• a CV.

• a motivation letter in English relating your track record to the specifics of the call.

• names and contacts of 2 referees.

Additional information

HT offers a highly collaborative, international culture. The working language at HT is English. HT will foster top quality, interdisciplinary research by promoting a vibrant environment consisting of independent research groups with access to outstanding graduate students, postdoctoral fellows and core facilities.

HT is an inclusive, equal opportunity employer offering attractive conditions and benefits appropriate to a leading, internationally competitive, research organization and seeks to promote a collegial and open atmosphere. The compensation package granted will be internationally competitive and comprise pension scheme, medical and other social benefits.

Special consideration will be given to candidates who are part of the protected categories list, according to L. 68/99.

Number of positions offered: 1

Contract offered: CCNL Chimico Farmaceutico, Fixed-term 4 years – employee level.

The position is based in Milan.