-
Matches from Samples FastQ
-
Reference Genome Matches
-
BH-TSNE of Bacterial Reference DNA
Inspiration
Our inspiration was to discover patterns within human DNA bacteria contamination by use of supervised and unsupervised machine learning.
What it does
Our techniques utilize various artificial intelligence techniques like machine learning to identify species based on signatures and probabilities.
How we built it
Using open source languages like perl and python, computing environments including AWS, open-source genomics tools like ACDC and minimap, and machine learning and NLP techniques.
Challenges we ran into
The volume of reference files were prohibitively large for computing power in a short period of time.
Accomplishments that we're proud of
Successfully plotted clusters of simulated bacterial gene sequences and identified relative frequencies within given DNA to show the concentrations of various species.
What we learned
Genomics is a task.
What's next for WhiteCoat Innovations
Continue to apply emerging technologies to contribute to health challenges.
Built With
- anomaly-detection
- linear-decomposition
- machine-learning
- natural-language-processing
- non-linear-decomposition
Log in or sign up for Devpost to join the conversation.