Assisting Illumina Ventures portfolio companies with their data analysis strategy and implementation in AI, machine learning and computational biologyPerforming due diligence to evaluate investment opportunities, including in-depthassessment of scientific and computational capabilities of therapeutics, diagnostics, and omics tools startups at Seed and Series A stages.

Encoded Therapeutics is funded by world-class healthcare investors Venrock, Arch Venture Partners, Illumina Ventures, and Matrix Capital Management. At Encoded, we are combining the precision of gene regulation with the versatility of gene therapies.I provided strategic and technical leadership for a 40-person R&D team with four functional areas: i) Computational Biology: Machine learning (with diverse approaches, including convolutional neural networks) for NGS screens of RE potency and specificity, analysis of single cell RNASeq data for validating AAV performance in vivo, and pipeline development for automated analysis, data management and visualization of all NGS screens.ii) Technology Development: NGS screening and validation of AAV regulatory sequence elements for potency and specificity in various clinically relevant tissues and cell types, and NGS screening for host cell factors that enhance AAV production.iii) Process Development: development of upstream and downstream processes for our suspension based platform for producing GMP grade AAV material.iv) Analytical Methods: development and qualification of analytical methods for measuring the yield and quality of GMP grade AAV.

Adaptive is the leading company in immune cell receptor sequencing. We have numerous commercial products in three broad areas: research, diagnostics and therapeutics.I provided strategic and technical leadership for a 50-person R&D team located in both Seattle and South San Francisco:i) Minimal Residual Disease: Project Leader for a 30-person cross-disciplinary team that developed Adaptive’s NGS-based ClonoSeq MRD assay, the first and only FDA approved MRD assay for multiple myeloma and ALL. This assay has received Medicare coverage and is becoming the de-facto surrogate endpoint for multiple myeloma clinical trials. Gave numerous presentations on the technical and clinical performance of the assay to Pharma partners, academic collaborators and board members as well as at international conferences and workshops. ii) Computational Biology: Analysis of clinical and pre-clinical research data from internal and external Immunosequencing projects. These projects have led to numerous high-profile scientific publications in cancer immunotherapy, autoimmune diseases and hematological oncology.iii) Assay Development: Research and development of molecular assays for immune receptor sequencing of T and B cells. Released several novel assays with exquisite sensitivity, robustness, bias control and industry-leading capabilities.iv) Analysis Pipeline: Development of novel algorithms and software systems to analyze NGS data from proprietary immunosequencing assays. Computational support of assay development.

Strategic and technical leadership of a diverse team of 25 scientists and engineers covering four functional groups:i) Computational Biology: Analyzed clinical data from Sequenta’s proprietary Minimal Residual Disease (MRD) technology. Analyzed immune cell repertoires in immunotherapy & autoimmune diseases, including Ankylosing Spondylitis and Sequenta’s proprietary MIRA technology for elucidating the relationship between T-cell VDJ sequences and antigen recognition. Gave oral presentations at international conferences and workshops.ii) Sequence Analysis: Developed Sequenta’s sequence analysis pipeline based on open source high-performance computing frameworks. Invented and coded novel sequencing algorithms for ultra-deep sequencing and annotation of VDJ clonotypes. Supported assay development. iii) LIMS & LPL: Developed a Laboratory Information Management System to automate the CLIA & research lab processes. This fully integrated system handled incoming test orders from physicians, orchestrated lab processes and delivered customized clinical reports through a portal.iv) IT: Servers, network, Linux compute cluster and desktop support.

Led the development of algorithms and software tools for the Oncology Research Group at Affymetrix. This group was developing next generation products in DNA resequencing, methylation & copy number assessment. Developed custom machine learning algorithms (using techniques such as support vector machines) to reach unprecedented levels of accuracy and sensitivity (< 10E-7 false positive rate with > 80% sensitivity) in resequencing that enabled the first pan-cancer map of somatic mutations.Created large sets of universal tag arrays using physical models of chip hybridization & cross-hybridization.

Led the algorithm development for ParAllele’s flagship products in DNA genotyping, copy number assessment & variation scanning. Developed high performance E/M algorithms with automated unsupervised clustering. Used physical modelling to adjust for probe saturation effects in copy number assessment. Led the implementation of algorithms embedded in applications shipped to customers. Developed numerous in-house software tools that enabled lab scientists to analyze their data daily.

Principal Architect and Technical Leader for Viaken’s Genomics Discovery Platform, an extensible framework for accessing, analysing and visualising genomic data. This new product comprised an extensible data warehouse accessed via a J2EE server from either a browser-based thin client or a Java Swing application.

Designed in UML and implemented in Java components of the Synomics Alliance Framework. This framework enabled pharmaceutical companies to integrate multiple data sources and processes into a single coherent distributed application. It was based on object oriented technology, relational databases, and Java frameworks such as J2EE.

Led research groups that developed software for astro-particle physics experiments, with emphasis on machine learning, data analysis, real-time data acquisition and Monte Carlo simulations. Principal Investigator for UK team in the ANTARES neutrino astronomy international project. Gave graduate & undergraduate lectures in physics. Thesis Advisor for a successful Ph.D. student.

Developed machine learning algorithms (using techniques such as simulated annealing and genetic algorithms) and Monte Carlo simulations for data analysis of solar neutrino physics experiments (SNO international collaboration).

Developed methods for assaying radioactive backgrounds in solar neutrino detectors. Reached a sensitivity of 1 part in 10E16 by concentrating ~100 atoms of 212Pb from 100 tons of water down to 10 ml in < 6 hours and measuring the quasi-synchronous (300 ns delay) beta decay of 212Bi and alpha decay of 212Po.Developed machine learning algorithms and Monte Carlo simulations for data analysis of solar neutrino physics experiments (SNO international collaboration).Gave undergraduate tutorials and lab supervision in physics.Doctoral Advisor for a successful Ph.D. student.