The value of machine learning methods in genome research has been growing nowadays. However, researchers often have to use obsolete software for scientific work. This is due to the unavailability of free access to the latest models. But soon, we can see this changing with the availability of free open access repository called “Kipoi”.
This repository allows an easy exchange of machine learning models involved in genomic research. The repository was created by Julien Gagneur, Assistant Professor of Computational Biology at the TUM, in collaboration with other researchers.
Researchers view about Kipoi
“What makes Kipoi special is that it provides free access to machine learning models that have already been trained,” says Julien Gagneur.
“What we are doing with Kipoi is not just sharing data and software, but sharing models and algorithms that are already trained on the most relevant data. These models are ready to use because all the cumbersome work of applying them to data has already been done,” says Anshul Kundaje, Assistant Professor (Stanford University)
Recently a study was published in the Nature Biotechnology which showcases the abundant benefits it will bring. The study shows that Kipoi will help in improving the exchange of models among genomics community thus paving the way for better genomic research.
Fast and simplified process
Kipoi can help the researcher to perform transfer learning. Since the repository already has trained models for particular datasets, it can be used for performing a similar task. This way it will be capable of learning the task at a faster pace than the earlier method. Kipoi also outperforms by simplifying the process of feeding data into the stored models in the repository. The entire process is simplified to the extent that a person with no experience in machine learning can use the repository efficiently.
In-depth analysis of genetic conditions
The use of Kipoi can lead to significant improvement in the easy identification of the genetic cause of diseases. This is because, Kipoi works with models that link genotype and phenotype, helping in assessing the genetic condition in depth.
“Kipoi puts the latest deep learning models trained on massive genomics data at the fingertips of clinical researchers,” says Julien Gagneur. “This provides very exciting opportunities to understand individual genomes, for instance, to pinpoint genetic variants causing diseases or to interpret mutations occurring in tumors.”
The advent of digitization has paved the way for new technology, machine learning is one among them. The use of machine learning to develop Kipoi repository opens the door for new research in the field of genomics.