EndToEndML - An Open-Source End-to-End Pipeline for Machine Learning Applications

Aug 3, 2026

Upload by Maeve

Artificial intelligence techniques are widely used in the life sciences, yet applying them to complex biological data is slowed by the steep learning curve for computing tools and programming. The approach proposed here delivers an open-source, user-friendly web pipeline that preprocesses, trains, evaluates, and visualizes machine learning models without manual coding. By integrating traditional ML and deep neural networks with visual analytics, it supports recognition, classification, clustering, and prediction across multimodal, multisensor datasets for drug discovery, pathogen classification, and medical diagnostics.

LanguageEnglish

CategoryTechnology

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Need for accessible, interpretable AI in life sciences

Challenges of fragmented ML tools and integration

EndToEndML as an end-to-end solution

EndToEndML: An Open-Source End-to-End Pipeline for Machine Learning Applications arXiv :2403 . 18203v1 [ cs .AI] 27 Mar 2024 1st Nisha Pillai Mississippi State University Mississippi State, USA [pillai@cse.msstate.edu](pillai@cse.msstate.edu)[ ](pillai@cse.msstate.edu)*Corresponding author 4th Ganga Gireesan Mississippi State University Mississippi State, USA [gg733@msstate.edu](gg733@msstate.edu) 2nd Athish Ram Das Mississippi State University Mississippi State, USA [ar2903@msstate.edu](ar2903@msstate.edu) 3rd Moses Ayoola Mississippi State University Mississippi State, USA[mba185@msstate.edu](mba185@msstate.edu) 5th Bindu Nanduri Mississippi State University Mississippi State, USA [bnanduri@cvm.msstate.edu](bnanduri@cvm.msstate.edu) 6th Mahalingam Ramkumar Mississippi State University Mississippi State, USA [ramkumar@cse.msstate.edu](ramkumar@cse.msstate.edu) Abstract—Artificial intelligence (AI) techniques are widely applied in the life sciences. However, applying innovative AI techniques to understand and deconvolute biological complexity is hindered by the learning curve for life science scientists to understand and use computing languages. An open-source, user-friendly interface for AI models, that does not require programming skills to analyze complex biological data will be extremely valuable to the bioinformatics community. With easy access to different sequencing technologies and increased interest in different ‘omics’ studies, the number of biological datasets being generated has increased and analyzing these highthroughput datasets is computationally demanding. The majority of AI libraries today require advanced programming skills as well as machine learning, data preprocessing, and visualization skills. In this research, we propose a web-based end-to-end pipeline that is capable of preprocessing, training, evaluating, and visualizing machine learning (ML) models without manual intervention or coding expertise. By integrating traditional machine learning and deep neural network models with visualizations, our library assists in recognizing, classifying, clustering, and predicting a wide range of multi-modal, multi-sensor datasets, including images, languages, and one-dimensional numerical data, for drug discovery, pathogen classification, and medical diagnostics. Index Terms—machine learning, bioinformatics, neural networks I. INTRODUCTION Biomedical research communities are increasingly in need of more accessible and interpretable artificial intelligence (AI) tools to extract insights from increasingly large and complex datasets. Performing machine learning on multifaceted biological data currently requires specialized programming expertise, limiting its adoption to bioinformaticians and computational biologists alone. The development of an open-source, userfriendly interface that enables researchers without coding skills to apply AI algorithms to data such as the microbiome related to soil, livestock, plant health, and other domains could dramatically expand the user base. By abstracting away the coding complexity through an intuitive graphical interface and prebuilt analysis workflows, more biologists could utilize the pre- dictive and pattern recognition capabilities of AI. This could accelerate discovery and the generation of hypotheses from high-dimensional biological data. Additional capabilities like interactive visualizations could also improve user understanding of the patterns and relationships detected by the models. An open-source platform would facilitate collaboration and allow customization of the built-in AI tools as methodologies continue to evolve. Overall, democratizing AI would equip a wider range of biologists to derive clinically and biologically relevant insights from increasingly complex and multi-modal data. This can profoundly enhance and accelerate biomedical research. Nowadays, there exists a wide array of machine learning libraries and frameworks, such that ev

FAQs

What problem does EndToEndML address for life science researchers?

It addresses the difficulty life science researchers face when applying ML to complex biological datasets due to coding and pipeline complexity requirements.

What does the EndToEndML web pipeline support end to end?

It supports preprocessing, training, evaluation, and visualization of ML models with automated workflows and no manual coding intervention.

How does EndToEndML help with model understanding and assessment?

It benchmarks performance using built-in evaluation metrics and provides interactive visualizations to explain patterns and relationships learned by models.

Aug 3, 2026

Upload by Maeve