Jesse W. Collins, Ph.D. Email & Phone Number

Boston, Ma, Us

Boston, Ma, Us

Developing our CANDID™ suite of AI/ML models for ADMET property prediction

Boston, Ma, Us

Boston, Ma, Us

Boston, Ma, Us

Indianapolis, Indiana, Us

React and React Native

ReactJS and nodejs software development for Coactive/CAI.During this time I also completed online coursework in SQL and node-SQL.

- Resumed work on a writing therapy nodejs app prototype- Submitted an 'ok'd' project pitch for NSF SBIR/STTR funding in "maker manufacturing"

Cambridge, Massachusetts, Us

- trained and tested a binary classifier and BERT-based AI model to infer gene-gene interactions from literature; tools included scikit learn and spacy (NLP)- made a webapp to analyze open source cell line CRISPR dependency map data from the Broad Institute Achilles project; tools included React/Python/MongoDB/Pandas/scipy and nference knowledge graph APIs, statistical measures used the hypergeometric distribution- calculated using HPC methods cosine similarities of cell types and tissue types based on RNA expression vectors across 18,000 genes, similarly cell line CRISPR dependency similarities- client interfacing work included projects with Janssen (J&J) and Syngenta, calculated PCA (Principle Component Analysis) plots for the latter NLP project

I make microfluidic junctions in a clear rubber (PDMS), and use them to make collagen drops containing neocortical progenitor stem cell spheroids (human iPSC-derived). The collagen gels around the cells to cue their differentiation to neurons. The resulting brain organoids could be used for studying basic (ie, "where are memories located?") and applied (ie, screening drugs against alzheimer's phenotypes) questions in neuroscience. With our dropmaking devices, we make hundreds of collagen drops per minute, about 100 times faster than when the gels are made by hand pipettor.

High throughput assay development for directed evolution of polymerases in microfluidic drops for neuroscience applications.- via a dilution series, proved we could detect drops containing a single molecule of polymerase DNA- worked with in vitro transcription and translation reagents, and a GFP reporter assay- screened drops at .2 khz rates

During this time I worked on converting 2 chapters from my thesis into journal articles, on software projects with friends/family and colleagues (see projects), and on looking for jobs in industry and science.

Cambridge, Massachusetts, Us

What if, after gently shaking a box of jigsaw puzzle pieces, you found the puzzle had completed itself? I studied analogous systems: proteins and DNA-coated microspheres. The point was to understand the physics of biological molecules. How do proteins fold and organize? How does life as we know it emerge from the seemingly random motion of petite, perhaps purposeful parcels of matter?After first studying a protein, Dscam, we turned to DNA-coated microspheres, an experimental system that circumvented a few problems with studying proteins directly. Individual proteins are too small and fast to see well, but microspheres are large and slow enough to watch with an optical microscope. Proteins are also very complex, consisting of a large number of different amino acids, whose interactions are not fully understood nor controllable. We studied small numbers of microspheres, so we could make sense of all the different clusters they could possibly form. We controlled the interactions between microspheres by controlling the DNA sequences on them. Although proteins and DNA-coated microspheres are thus very different, we believe that proteins and microspheres are obey similar physics. Statistical mechanics can be applied to both systems.In these thesis experiments, we found that with careful control of the interactions between the building blocks, uncommonly symmetric structures self-assemble. Link to my thesis: https://dash.harvard.edu/handle/1/12274201

Cambridge, Massachusetts, Us

Co-developed lesson plans, graded quizzes and homeworkLead section for 20 studentsCo-administered homework help session for up to about 50 students at a timeCoursework including the following subjects: kinematics, acceleration, friction, force balances, torque, energy conservation

Cambridge, Ma, Us

Held office hours to help students individually and in small groups

Formulated a UV curable ferrofluid, optimized continuous droplet polymerization flow rates, made PDMS microfluidic devices, demonstrated “janus” microparticle polymerization via laminar coflowPerformed microscopy of microparticles and collaborated on a journal article (see publications)

Cambridge, Ma, Us

Made SQUID (super-conducting quantum interference device) and ESR (electron spin resonance)measurements of colloidal CdSe quantum dots down to 4K, found an ESR peak at 39K.

Tested local water samples for presence of microorganisms, mixed media, cleaned and sterilized glassware.

Doctor Of Philosophy (Phd), Applied Physics

Ms, Applied Physics

Bachelor Of Science (Bs), Physics

Bachelor Of Science (Bs), Literature

Education record

Education record

Education record