Deep learning is pushing the frontiers of biology. Here’s how

Next generation sequencing has allowed major speed gains with the use of massive parallel sequencing.

Healthcare providers are extensively using AI and ML to deliver personalised treatments. Lately, ML has proved instrumental in driving clinical trial efficiency and expediting the drug discovery process.

Scientists are now tapping deep learning algorithms for cellular image classification, drug discovery, genome analysis and also to link image and genome data with EMR (Electronic Medical Records). In computational biology, deep learning is applied on high-throughput biological data to make sense of high-dimensional datasets.

ML applications in biology

Next generation sequencing has allowed major speed gains with the use of massive parallel sequencing. Now, ML is applied to identify gene coding regions in a genome. Such gene prediction tools have been a huge improvement on typical homology-based sequence searches.

In proteomics, the use of ML has increased the structure prediction accuracy from 70% to over 80%. In addition, text-mining of secondary databases and multiple journal articles has helped identify novel drug targets.

ML tools in biology

Deep learning methods such as recurrent neural network (RNN), deep neural network (DNN), deep autoencoder (DA), convolution neural network (CNN), and deep Boltzman machine (DBM) have umpteen use cases in biology. For instance, DNN plays a critical role in identifying potential biomarkers from proteome and genome data. DeepCpG is a CNN application used to predict DNA methylation states in single cells. It has delivered accurate results compared to existing methods while vetting five types of methylation data.

In 2005, computational biologist Anne Carpenter developed an open-source software, CellProfiler, to quantitatively measure individual features such as the number of fluorescent cells in a microscopy field. 

Verily life science and Google combinedly designed DeepVariant to mine genome data. It predicts the common type of genetic variation accurately compared to conventional tools. 

Deep learning has extensive applications in the drug discovery process. Atomwise has developed an algorithm to represent molecules in 3D pixels–it helps visualise the 3D structure of proteins and small molecules with atomic precision. Further, the algorithm uses these features to predict small molecules that interact with a given protein.

According to Dr Melanie Matheu, founder at Prellis Biologics, Inc. the new generation of therapeutics entering drug pipelines empowered by AI screening for selecting targets is poised to minimise clinical trial failure rates for small molecules by as much as 86%.

AI’s cost efficiency

AI algorithms can crunch insane amounts of data to speed up the drug discovery process. The use of AI/ML is expected to bring down the existing 90% failure rate in Pharmaceutical manufacturing.

In the Human Genome Project, data analytics and ML algorithms have helped reduce the cost of genome sequencing from USD 2.7 billion to less than $300. Now, bioengineers are able to perform extensive screening of trial participants with the use of these cutting edge technologies. In the near future, healthcare providers will be able to design treatments that work at genetic and molecular levels.

More Great AIM Stories

Meeta Ramnani
Meeta’s interest lies in finding out real practical applications of technology. At AIM, she writes stories that question the new inventions and the need to develop them. She believes that technology has and will continue to change the world very fast and that it is no more ‘cool’ to be ‘old-school’. If people don’t update themselves with the technology, they will surely be left behind.

More Stories

OUR UPCOMING EVENTS

8th April | In-person Conference | Hotel Radisson Blue, Bangalore

Organized by Analytics India Magazine

View Event >>

30th Apr | Virtual conference

Organized by Analytics India Magazine

View Event >>

MORE FROM AIM
Yugesh Verma
All you need to know about Graph Embeddings

Embeddings can be the subgroups of a group, similarly, in graph theory embedding of a graph can be considered as a representation of a graph on a surface, where points of that surface are made up of vertices and arcs are made up of edges

Yugesh Verma
A beginner’s guide to Spatio-Temporal graph neural networks

Spatio-temporal graphs are made of static structures and time-varying features, and such information in a graph requires a neural network that can deal with time-varying features of the graph. Neural networks which are developed to deal with time-varying features of the graph can be considered as Spatio-temporal graph neural networks. 

Yugesh Verma
A guide to explainable named entity recognition

Named entity recognition (NER) is difficult to understand how the process of NER worked in the background or how the process is behaving with the data, it needs more explainability. we can make it more explainable.

3 Ways to Join our Community

Discord Server

Stay Connected with a larger ecosystem of data science and ML Professionals

Telegram Channel

Discover special offers, top stories, upcoming events, and more.

Subscribe to our newsletter

Get the latest updates from AIM