DeepSmallCas9 Deep Learning Model for Small Cas9 Activity Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Selecting the optimal small Cas9 for specific target sequences is challenging due to the complexity of their activities and specificities, and existing computational models do not effectively predict activities at both matched and mismatched target sequences.
Innovation Solution
A deep learning-based system, DeepSmallCas9, is developed to predict the activities of small Cas9s by learning the relationship between guide sequences, target sequences, and features affecting activity, enabling the selection of appropriate small Cas9 and sgRNA for genome editing projects, including at mismatched target sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If small Cas9 orthologues and variants are used for delivery, then delivery efficiency is improved, but activity prediction accuracy deteriorates
Solution Approach 1:
The patent applies parameter changes by training the deep learning model on diverse small Cas9 orthologues and variants with different structural parameters, enabling the model to accurately predict activities across various Cas9 types while maintaining delivery efficiency benefits
Solution Approach 2:
The patent uses computational modeling to create virtual representations of small Cas9 activities, allowing researchers to predict and compare activities of different Cas9 variants without extensive wet-lab experimentation, thus improving prediction accuracy while maintaining delivery efficiency
2Loss of time
If deep learning models are trained on limited experimental data, then model training speed is improved, but prediction reliability deteriorates
Solution Approach 1:
The patent applies preliminary action by curating and preprocessing high-quality experimental data before model training, organizing it in a format that enables efficient learning while ensuring comprehensive coverage of small Cas9 activity patterns, thus achieving both fast training and reliable predictions
Solution Approach 2:
The patent creates a universal deep learning model that can predict activities across multiple small Cas9 orthologues and variants simultaneously, allowing the model to learn from diverse data sources and improve reliability through multi-functional prediction capabilities
Data Source
AI summary
A system for predicting an activity of small Cas9 using deep learning, including a sequence input unit receiving input data on a guide sequence and target sequence of small Cas9, a predictive model generator generating a small Cas9 activity predictive model by performing deep learning for learning a relationship between small Cas9 activity data obtained from the input data on the guide sequence and target sequence of small Cas9 received from the sequence input unit and features that affect small Cas9 activity, a candidate target sequence input unit receiving candidate target sequence of small Cas9, and an activity predictor predicting small Cas9 activity by applying candidate target sequence input in the candidate target sequence input unit to the predictive model generated in the predictive model generator.