Computational Botany: A Review of Machine Learning, IoT, and Bioinformatics in Plant Sciences

Abstract

The intersection of computer science and botany has catalyzed a paradigm shift in plant taxonomy, pathology, and agricultural management. This paper reviews the current literature on computational applications in plant sciences, focusing on three primary domains: Deep Learning for automated disease detection, Internet of Things (IoT) sensor networks for physiological monitoring, and high-performance computing in plant genomics. By synthesizing recent methodologies, this review identifies existing technological bottlenecks—such as dataset biases and processing latencies—and outlines future trajectories for integrating artificial intelligence into sustainable botanical research and precision agriculture.

Introduction

This research explores the integration of computer science and plant sciences (Computational Botany) to improve plant research, disease detection, agricultural productivity, and environmental monitoring. Traditional botanical research relies heavily on manual observation, laboratory experiments, and time-consuming data collection. However, increasing agricultural demands, climate change, and the growing complexity of biological data have created a need for advanced computational technologies.

The proposed research highlights three major technological areas: Artificial Intelligence (AI), the Internet of Things (IoT), and Bioinformatics.

1. Deep Learning and Computer Vision in Plant Pathology

Artificial intelligence and deep learning have transformed plant disease detection by enabling automated analysis of plant images. Convolutional Neural Networks (CNNs), YOLO, ResNet, VGG16, and Faster R-CNN are used to identify plant diseases, detect infected regions, classify plant species, and assess plant health.

Previous studies have demonstrated high disease-classification accuracy using datasets such as PlantVillage. Lightweight models such as MobileNet and EfficientNet can also be deployed on smartphones and agricultural drones for real-time disease detection.

However, challenges include limited dataset diversity, changing environmental conditions, poor performance in real-world field images, and the lack of interpretability in deep learning models. Explainable AI techniques such as Grad-CAM can help researchers understand which image features influence model predictions.

2. IoT and Sensor Networks for Microclimate Monitoring

IoT and Wireless Sensor Networks (WSNs) enable continuous monitoring of environmental conditions affecting plant growth and development.

Sensors are used to measure:

• Soil moisture and water availability.

• Temperature and relative humidity.

• Photosynthetically Active Radiation (PAR).

• Environmental conditions influencing plant growth and stress.

The collected data can be processed using cloud computing and machine-learning algorithms to predict drought stress, optimize irrigation, and improve precision agriculture.

Communication technologies such as LoRaWAN allow sensors to transmit information over long distances while consuming less power.

Major challenges include sensor degradation, inaccurate measurements, limited battery life, environmental damage, and the management of large volumes of sensor data. Machine-learning techniques such as Long Short-Term Memory (LSTM) networks can help analyze environmental time-series data and identify meaningful patterns.

3. Bioinformatics and Plant Genomics

Bioinformatics combines computer science, mathematics, and biological sciences to analyze complex plant genetic information.

Next-Generation Sequencing (NGS) and advanced sequencing technologies generate large volumes of genomic, transcriptomic, and proteomic data. Processing this information requires powerful computational algorithms and high-performance computing.

Important computational techniques include:

• De Bruijn Graph Algorithms: Used for assembling short DNA sequences into larger genomic sequences.

• Overlap-Layout-Consensus (OLC): Used for reconstructing genomes from long-read sequencing data.

• RNA-Seq Analysis: Used to study gene expression and understand plant responses to environmental stress.

• Machine Learning: Helps identify genes associated with drought tolerance, disease resistance, and other important plant characteristics.

These technologies support molecular breeding, genetic research, and the development of climate-resilient crop varieties.

4. Major Research Challenges

The literature review identifies several important challenges:

1. Limited availability of geographically diverse and accurately labelled plant datasets.

2. Difficulty in interpreting predictions generated by complex deep learning models.

3. Challenges in deploying computational models in remote and resource-constrained environments.

4. Sensor reliability, power consumption, and environmental degradation in IoT networks.

5. Increasing computational and storage requirements for genomic and multi-omics data.

6. Limited integration between biological research and advanced computer science technologies.

5. Expected Outcomes

The integration of AI, IoT, and bioinformatics is expected to improve plant disease detection, automated phenotyping, environmental monitoring, and genomic analysis.

It can support:

• Early and accurate identification of plant diseases.

• Real-time monitoring of environmental conditions.

• Smart irrigation and precision agriculture.

• Improved understanding of plant responses to environmental stress.

• Identification of genes responsible for desirable agricultural traits.

• Development of climate-resilient and high-productivity crop varieties.

Conclusion

The synthesis of computer science and botany represents a fundamental paradigm shift in how humanity understands, monitors, and manipulates plant life. As this literature review demonstrates, computational tools are no longer merely auxiliary accessories to traditional botanical research; they have become the foundational infrastructure required to navigate the complexities of modern biology. The transition from manual, empirical observation to high-throughput, data-driven analysis is rapidly accelerating across all scales of plant science. At the organismal level, deep learning and computer vision architectures have revolutionised plant pathology, moving diagnostics from the laboratory directly to the field via edge computing. At the ecological level, the deployment of IoT sensor networks and LoRaWAN protocols has provided an unprecedented, real-time window into ecophysiology, enabling precision agriculture to proactively respond to micro climate fluctuations. Simultaneously, at the molecular level, advanced bio-informatics algorithms and graph theory are untangling the immense complexity of polyploid genomes and mapping dynamic transcriptomic responses to environmental stress. However, realising the full potential of computational botany requires addressing several critical bottlenecks. The literature consistently highlights the need to overcome the \"domain shift\" in field-deployed machine learning, the demand for Explainable AI to restore biological interpretability to neural networks, and the necessity of robust computational infrastructure to support the incoming wave of multi-omics and pan-genomic data. Ultimately, the future of global food security, climate change adaptation, and sustainable agriculture depends on the continued convergence of these disciplines. By fostering interdisciplinary collaboration and developing scalable, accessible computational frameworks, researchers can transform the exponential influx of biological data into actionable insights, ensuring the resilience of plant ecosystems in a rapidly changing world.

References

[1] F. Fiorani and U. Schurr, “Future Scenarios for Plant Phenotyping,” Annu. Rev. Plant Biol., vol. 64, pp. 267–291, 2013, doi: 10.1146/annurev-arplant-050312-120137. [2] A. Kamilaris and F. X. Prenafeta-Boldú, “Deep learning in agriculture: A survey,” Comput. Electron. Agric., vol. 147, pp. 70–90, Apr. 2018, doi: 10.1016/j.compag.2018.02.016. [3] A. Singh, B. Ganapathysubramanian, A. K. Singh, and S. Sarkar, “Machine Learning for High-Throughput Stress Phenotyping in Plants,” Trends Plant Sci., vol. 21, no. 2, pp. 110–124, Feb. 2016, doi: 10.1016/j.tplants.2015.10.015. [4] A. Tzounis, N. Katsoulas, T. Bartzanas, and C. Kittas, “Internet of Things in agriculture, recent advances and future challenges,” Biosyst. Eng., vol. 164, pp. 31–48, Dec. 2017, doi: 10.1016/j.biosystemseng.2017.09.007. [5] “Genomics and Bioinformatics Resources for Crop Improvement | Plant and Cell Physiology | Oxford Academic.” Accessed: Oct. 27, 2026. [Online]. Available: https://academic.oup.com/pcp/article/51/4/497/1828103 [6] S. Savary, L. Willocquet, S. J. Pethybridge, P. Esker, N. McRoberts, and A. Nelson, “The global burden of pathogens and pests on major food crops,” Nat. Ecol. Evol., vol. 3, no. 3, pp. 430–439, Mar. 2019, doi: 10.1038/s41559-018-0793-y. [7] S. P. Mohanty, D. P. Hughes, and M. Salathé, “Using Deep Learning for Image-Based Plant Disease Detection,” Front. Plant Sci., vol. 7, Sep. 2016, doi: 10.3389/fpls.2016.01419. [8] J. G. Arnal Barbedo, “Plant disease identification from individual lesions and spots using deep learning,” Biosyst. Eng., vol. 180, pp. 96–107, Apr. 2019, doi: 10.1016/j.biosystemseng.2019.02.002. [9] J. Chen, J. Chen, D. Zhang, Y. Sun, and Y. A. Nanehkaran, “Using deep transfer learning for image-based plant disease identification,” Comput. Electron. Agric., vol. 173, p. 105393, Jun. 2020, doi: 10.1016/j.compag.2020.105393. [10] K. P. Ferentinos, “Deep learning models for plant disease detection and diagnosis,” Comput. Electron. Agric., vol. 145, pp. 311–318, Feb. 2018, doi: 10.1016/j.compag.2018.01.009. [11] “How Convolutional Neural Networks Diagnose Plant Disease | Plant Phenomics.” Accessed: Oct. 27, 2026. [Online]. Available: https://spj.science.org/doi/full/10.34133/2019/9237136 [12] E. Navarro, N. Costa, and A. Pereira, “A Systematic Review of IoT Solutions for Smart Farming,” Sensors, vol. 20, no. 15, p. 4231, Aug. 2020, doi: 10.3390/s20154231. [13] P. P. Ray, “Internet of things for smart agriculture: Technologies, practices and future direction,” J. Ambient Intell. Smart Environ., vol. 9, no. 4, pp. 395–420, Jun. 2017, doi: 10.3233/AIS-170440. [14] M. O. Yerebakan and B. Hu, “Human–Robot Collaboration in Modern Agriculture: A Review of the Current Research Landscape,” Adv. Intell. Syst., vol. 6, no. 7, p. 2300823, 2024, doi: 10.1002/aisy.202300823. [15] T. P. Michael and S. Jackson, “The First 50 Plant Genomes,” Plant Genome, vol. 6, no. 2, p. plantgenome2013.03.0001in, 2013, doi: 10.3835/plantgenome2013.03.0001in. [16] R. Li et al., “De novo assembly of human genomes with massively parallel short read sequencing,” Genome Res., vol. 20, no. 2, pp. 265–272, Feb. 2010, doi: 10.1101/gr.097261.109. [17] A. Conesa et al., “A survey of best practices for RNA-seq data analysis,” Genome Biol., vol. 17, no. 1, p. 13, Jan. 2016, doi: 10.1186/s13059-016-0881-8. [18] P. E. Bayer, A. A. Golicz, A. Scheben, J. Batley, and D. Edwards, “Plant pan-genomes are the new reference,” Nat. Plants, vol. 6, no. 8, pp. 914–920, Aug. 2020, doi: 10.1038/s41477-020-0733-0.

Copyright

Copyright © 2026 Nishchal Sharma, Dr. Harish Chand. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.