In this article, we will discuss the application of AI in Bioinformatics. The outline of the topic will be
Introduction
Preface of Molecular Biology
Basics of AI
Examples of AI applications in Bioinformatics
Conclusion
Before the introduction, I want readers to think , why are we different ?Why are we similar? To find the answer, you have to read the entire article!!!
Introduction:
Bioinformatics is a field of study that involves various computational aspects to extract information from biological data. It includes data collection, data storage, data retrieval, data manipulation, and modeling to show the visualization and prediction through algorithms and software.
Nowadays, AI is gaining more attention towards Biological aspects. It predicts a lot of applications that relate to Biological data.AI will help to analyze large datasets and give the solution quickly. For instance: Now, we are suffering from the COVID crisis. Normally, the drug will take a lot of time to process. With the help of AI, we were able to get the drug quickly, we saved a lot of people from this tragedy. Another example is individualized personalized medicine for cancer patients or other disease patients. It finds the early stage of cancer and finds the treatment quickly. In 2030, AI will improve a lot of aspects that relate to Biomedical research. In this article, I will explain the basics of Biology, AI with example articles that provide insights into Bioinformatics and Artificial intelligence.
Overview of Biology:
The above picture will clearly explain the evolution of ;
Genome->Chromosome->DNA->RNA->Protein->Metabolites
There are a lot of sequencing approaches for each steps.
Genome-Genome sequence/variation
Chromosome-Chromatin structure/Gene regulation/epigenetic studies
RNA-Transcriptomics studies/gene expression analysis
Protein-Proteomic approaches of finding the proteins using a lot of proteomic techniques like Mass-spectrometry.
Preface of Molecular Biology:
The central dogma is the major process and study of Molecular Biology. I asked the question earlier , why are we different? Why are we similar? The answer would be DNA. What is DNA?
DNA is the heredity of genetic material and is known as Deoxyribonucleic acid. DNA has 4 nucleotides such as A, T, G, C.DNA is double-stranded. The first phase of the central dogma, converting DNA to RNA by Transcription. It is a process of making RNA copies of gene sequences. This particular copy is called mRNA(messenger RNA).RNA is a single-stranded molecule. RNA has 4 nucleotides such as A, U, G, C. The new mRNA copies of the genes are the major protein synthesis during the process of translation .mRNA will encode the result of proteins. For the detailed explanation of central dogma, you can read it here[1]. Majorly, the problem will come from the regulation of transcription and translation. Systems Biology research is trending with AI aspects too.
Basics of AI resolves Bioinformatics problems:
The simulation of human intelligence processes by computers is known as artificial intelligence. There are special features of AI such as NLP(Natural language processing), Speech recognition and object detection.AI has two main things ;Machine learning and Deep learning. The below workflow will explain clearly about these three concepts.[2].
Why do we need Biomedical research? Why should we focus on AI? What are the benefits we require from AI in Biomedical? The below workflow will explain the need for AI in Bioinformatics:[5]
Biomedical research is important to finding cures for diseases and finding a lot of interesting things in the use of AI. In the healthcare system, now health data and AI is inseparable.
The above workflow explains how medical/research experts deal with AI in health systems. First, we have a lot of data and we need to analyze the data with statistical inference. We need to design the experiments according to the import of the data to understand AI. There are a lot of advantages of AI in making predictions with accuracy, understanding our ideas, and implement them accordingly, which makes the work really faster. It chooses the particular method to analyze the data.
AI will solve Big problems. There are a lot of completed projects with the help of AI such as the Human genome project,1000 genome project, ENCODE, TCGA, and so on
There are some recent approaches of AI in Bioinformatics:[6]-Table summary
Examples of applications of AI in Bioinformatics(Recent trends):
1.Artificial intelligence in COVID-19 drug repurposing:
This paper explains how AI models can work on finding the drug repurposing approach to treat COVID-19[7].
2.Application of Machine Learning to Proteomics Data: Classification and Biomarker Identification in Postgenomics Biology:
This paper explains the importance of AI in proteomic approaches to identify the biomarker for a particular disease within a short time[8].
There are a lot of papers out there that discuss the importance of AI in Biomedical research. I just highlighted a few papers.
The next example is one of my collaborative projects with my group members, when I was a project associate at IIT Madras. The paper was published in biorxiv.
Title of the paper: Predicting cross-tissue hormone-gene relations using balanced word embeddings: [9]
Using the Biomedical literature, Our model predicts the relationship of Hormone-gene associations. Based on the Bioinformatics validations, we found evidence that our model predicts very accurate results.
Conclusion:
In this article, I have explained the integration of Bioinformatics and Artificial intelligence. I have mentioned clear examples of how we can implement biological data into AI. The aim of this article is to understand the readers to know the ongoing research field.
References:
1. CRICK, F. Central Dogma of Molecular Biology. Nature 227, 561–563 (1970). https://doi.org/10.1038/227561a0
2. C. Williams, "A Brief Introduction To Artificial Intelligence," Proceedings OCEANS '83, 1983, pp. 94-99, doi: 10.1109/OCEANS.1983.1152096.
4. All reference pictures are taken from Google resources.
5. AI applications in Bioinformatics and precision medicine by Dr.Yufei Huang
6. Adapted from: Kun-Hsing Yu, Andrew L Beam, and Isaac S Kohane. 2018. “Artificial intelligence in healthcare.” Nature Biomedical Engineering, 2, 10, Pp. 719.
7. Artificial intelligence in COVID-19 drug repurposingZhou et al. - The Lancet Digital Health – 2020
8. Swan AL, Mobasheri A, Allaway D, Liddell S, Bacardit J. Application of machine learning to proteomics data: classification and biomarker identification in postgenomics biology. OMICS. 2013;17(12):595-610. doi:10.1089/omi.2013.0017
9. Predicting cross-tissue hormone-gene relations using balanced word embeddings
Jadhav et al. - 2021