Related Experiment Video
Updated: Jun 12, 2025
08:51
Identification of Mouse and Human Antibody Repertoires by Next-Generation Sequencing
Published on: March 15, 2019
12.3K
Harmonizing immune cell sequences for computational analysis with large language models
Areej Alsaafin1, Hamid R Tizhoosh1
1Department of Artificial Intelligence & Informatics, KIMIA Lab, Mayo Clinic, Rochester, MN, 55905, United States.
Biology Methods & Protocols
|September 18, 2024
Summary
Seqwash harmonizes immune cell sequences for large language models (LLMs), improving data quality. This standardization enhances downstream task performance in sequencing data analysis.
Area of Science:
- Bioinformatics
- Computational Biology
- Immunoinformatics
Background:
- Immune cell sequencing generates complex data.
- Standardizing sequencing profiles is crucial for accurate analysis.
- Large language models (LLMs) offer potential for pattern recognition in biological data.
Purpose of the Study:
- To introduce Seqwash (SEQuence Weighted Alignment for Sorting and Harmonization), an algorithm for processing sequencing profiles.
- To harmonize immune cell sequences into a unified representation for LLM analysis.
- To improve feature quality and downstream task performance using standardized sequencing data.
Main Methods:
- Development of the Seqwash algorithm.
- Application of Seqwash to immune cell sequencing data.
- Utilizing LLMs for pattern embedding and information filtering.
Main Results:
- Seqwash effectively harmonizes immune cell sequencing profiles.
- Standardized profiles exhibit improved feature quality.
- Enhanced performance observed in both supervised and unsupervised downstream sequencing data tasks.
Conclusions:
- Seqwash is an effective tool for standardizing immune cell sequencing data.
- The algorithm empowers LLMs to identify meaningful patterns in biological sequences.
- Seqwash facilitates improved downstream analysis of sequencing data.
