Abstract
In this research paper, we focus on designing and developing a fully automated gene regulation from cancerous cell heterogeneity using advanced machine learning techniques. There are several modern technologies developed to make DNA sequencing easier and cheaper. Among them, gene regulation produces the longest read sequences and the lengths of the reads are growing day by day. Machine learning technique like Support Vector Machine (SVM) is developed to align these gene sequences. Every technique faced some challenges, but facing the desired challenges introduce some new challenges on the other side. So, no one tool is perfect for every work. The SVM technique is a new aligner tool that does a tradeoff and performs better from different aspects. For the model with the best generator loss, an average maximum validation accuracy of 91.29% was achieved. The gene regulation with SVM is like a mini-map that takes a few times more memory to index the whole genome of a reference sequence. The single-cell data are the main target of SVM. It is shown that it would help the SVM and similar techniques to align better with long insertions and deletions of single-cell gene regulation. Single-cell data is run against the well-known reference sequence and a randomly generated synthetic reference.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have