Background
The Department of Plant Sciences at the University of Colombo needs to expand its high-performance computing facilities to conduct advanced research. In 2021, the Department received two high-performing computers as a generous donation from CUFSAA. These were effectively utilized, particularly by Bioinformatics Honours degree undergraduates for their final-year research projects. Notably, this facility enabled several students to conduct research of publication quality, leading to papers in SCI Expanded–indexed journals, a remarkable achievement at the undergraduate level, made possible through CUFSAA’s generous support.
In 2023, the annual student intake for the Bioinformatics Honours degree program was increased from 8 to 10 due to high demand among students. However, with only two high-performance computers available, students often experience delays due to the need to pre-book the facility. This significantly limits the time they can dedicate to analysis. The increasing computational demand of modern research projects has made the existing resources insufficient.
At present, the limited number of high-performance computing (HPC) facilities continues to restrict their capacity to conduct advanced research in genomics, transcriptomics, proteomics, and systems biology. Many current projects require processing large datasets, often ranging from 100 to 300 GB, generated by next-generation sequencing (NGS). Although the currently available HPCs have enabled high-throughput research, they lack the necessary memory and GPU capacity, making it impossible to run full pipelines or complete analyses within a reasonable timeframe.
A major limitation arises from the need for GPU acceleration and sufficient VRAM to support molecular dynamics simulations, protein structure modelling and predictions, and deep learning-based applications. Modern workflows, such as molecular modelling using CUDA-enabled software or machine learning models for functional genomics, require high GPU throughput alongside balanced CPU performance to avoid bottlenecks.
For instance, some research projects require an extensive amount of computing resources to train their deep learning models. The primary reason is the scale of the projects; the project for Protein Function Prediction for Chordate Proteins has a huge dataset comprising multi-modal data (structure, sequence, and network), amounting to over 100GB of space. In addition, to process this, the model would need around 20 – 30 GB of allocated VRAM if the model has to run on recommended parameters, which will capture the biological patterns in the data given.
Similarly, parallel processing of large-scale NGS analyses, such as genome assembly, variant calling, and transcript quantification, cannot be achieved with the current infrastructure. In fact, the lack of such resources has forced many research students to rely on external resources that are either only available through a waiting list or require payment.
Consequently, research students run jobs sequentially, prolonging analyses from hours into weeks, hindering the progress and delaying projects of other research students who are relying on these HPCs as well. This in turn limits the scope of research questions that can be addressed.
In addition to computational capacity, the lack of adequate storage infrastructure prevents us from managing and analyzing datasets of this scale efficiently. Moreover, many students store the required datasets for analysis in a given HPC, substantially diminishing the available storage space. Furthermore, analysis of NGS data routinely requires both fast storage for active computation and high-capacity storage for intermediate and final results. Without this, researchers face constant trade-offs between what data can be processed and what must be discarded, further limiting the reproducibility and reliability of their work.
The addition of two new high-performance computers will:
- Ensure equitable access for students and reduce waiting times.
- Allow students to perform advanced analyses, enhancing the quality of their research outputs.
- Strengthen the Department’s capacity to provide practical training in state-of-the-art bioinformatics.
- Directly benefit both undergraduates and postgraduates, as well as young academic staff members, by providing the computational power necessary to support advanced projects and collaborations with international laboratories.
This investment will directly enhance student learning, research capacity, and the faculty’s reputation as a leader in bioinformatics education and research in Sri Lanka.
By providing additional computing power, the Department can:
- Support complex and large-scale analyses (genome assembly, transcriptome mapping, SNP calling, phylogenetics, protein modeling).
- Enhance the quality of student projects, increasing their chances of publishing in peer-reviewed journals.
- Improve employability and postgraduate opportunities, as students will be trained on computational skills highly sought in both academia and industry.
- Provide resources for postgraduate research, thereby strengthening the Department’s research profile.
Proposed specifications
The Department wishes to request support for the purchase of two high-performance computers to meet the current requirements. The specifications and the estimated cost for a single high-performance computer are given in the table below.
| Product Name | Description | Quantity |
| Intel Core i7-14700F Desktop Processor (33M Cache, up to 5.40 GHz) | CPU | 1 |
| DEEPCOOL AG400 PLUS 120MM CPU Air Cooler | Cooler | 1 |
| Gigabyte B760 GAMING X AX DDR5 ATX Motherboard (WiFi) | Motherboard | 1 |
| Gigabyte GeForce RTX 5090 WINDFORCE OC 32G | VGA | 1 |
| Gigabyte AORUS 7000E Gen4 2TB SSD | SSD | 1 |
| G skill 64GB (32×2) DDR5 5600MHz Memory | RAM | 1 |
| 1600W Power Supply | Power Supply | 1 |
| Workstation Case | Casing | 1 |
| PROLINK UPS PRO2000SFC | UPS and Power Surge Protector | 1 |
| Fantech K211 GO Office Keyboard | Keyboard | 1 |
| UGREEN Wired Optical Mouse – 90789 | Mouse | 1 |
| VIEWSONIC VA1903H 19″ TN MONITOR | Monitor | 1 |
| Total | LKR 1,690,950
(Approx US$ 5,154.00/per workstation |
On behalf of CUFSAA-NA, we kindly invite you to make a generous contribution toward this project and help us turn this vision into a reality. Your support will make a meaningful difference and bring us one step closer to achieving our goal.
Donate & Support
Donations can be made using the QR code for PayPal, or through Zelle by contacting the treasurer (sliyanarachchi@cufsaa.org) or by mailing a check payable to CUFSAA NA (Memo: Nanofabrication Core Facility) to:
Colombo University Faculty of Science Alumni Association
c/o Ms. Sandya Liyanarachchi
Treasurer – CUFSAA-NA
5870 Tara Hill Drive
Dublin, OH 43017
USA