Mirvie
About the start-up
Mirvie is shaping the future of pregnancy health by providing women, expecting parents and their doctors with an early detection window to intervene before unexpected pregnancy complications become a crisis. One in five pregnancies is impacted by complications that lead to lifelong health consequences for expecting parents and babies.
The proprietary Mirvie RNA platform uses a simple blood test to reveal vital information about a pregnancy’s unique biology and detect complications months before they occur.
The Opportunity
Multiple opportunities will be available depending on the candidate profile and interests, a few examples are listed below.
Project 1 - Computer Science/Bioinformatics
We are working with a relatively new lab instrument (digital droplet polymerase chain reaction, ddPCR) and the software released by the company is controlled via a graphical user interface only. We would like to develop an application that can read in the raw files, calculate thresholds for each of our assays, and automatically output results for further interpretation. This will require the candidate to get familiar (at least theoretically) with ddPCR, and understand the poison process that will govern the assay threshold. Output can initially be in the form of a text file or as time allows additional plotting functions can be added.
Automating more advanced analyses of the data to make a classification for disease outcome will be a natural extension of this project.
Project 2 – Bioinformatics
We currently have an automated pipeline built that runs in AWS using an orchestration software (NextFlow) and various bioinformatics tools. We would like to expand that pipeline to add several additional capabilities for example variant calling (from RNAseq so nontrivial), haplotype caller. This would require learning AWS compute infrastructure, NextFlow as well as selecting relevant tools and substantiating the choice of tools.
A more advanced extension of this project would be to build a drift detection algorithm, based on learning from existing QC data is it possible to predict if the main NGS assay is drifting.
Primary responsibilities
Depending on the project chosen the candidate can choose to lean closer to a lab-integration project (with the option to learn the wetlab workflow) or work closer with the software/bioinformatics team to develop key infrastructure.
The candidate will work independently supervised by an experienced bioinformatician/data scientist/software engineer as needed. The candidate will be expected to provide progress updates during weekly meetings and to present to the broader R&D team on occasion.
About you
The candidate should have experience with standard software tools such as version control, code testing, code documentation. Proficiency in a Linux based environment is expected.
Python (or R) programming knowledge is preferred, as is knowledge of workflow orchestration (e.g. make, NextFlow, snakemake) and Docker.
Pluses
Some experience with either machine learning models or bioinformatics is a plus.
Looking for
Computer Science or BioInformatics
Location
South San Francisco, California