🧬 biology

Computational Identification and Prioritization of Conserved HIV-1 p24 Peptide Candidates

This study presents a reproducible computational pipeline that analyzed 185 HIV-1 p24 sequences to identify and prioritize 21 highly conserved peptide candidates, including a top MHC-I epitope with 98.4% sequence coverage, for potential vaccine development.

Original authors: Sepideh Moafi

Published 2026-08-24

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Sepideh Moafi

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

For decades, the global fight against HIV-1 has faced a formidable obstacle: the virus changes its shape constantly. This genetic shapeshifting allows it to hide from the immune system and evade vaccines, making the creation of a protective shot one of the most difficult challenges in modern medicine. Scientists have long looked for a part of the virus that stays the same, a structural piece so essential to the virus's survival that it cannot afford to mutate. One such piece is the p24 protein, a rigid shell that holds the virus together. Because this shell must maintain a specific form to function, it changes very little across different strains of the virus. If researchers can find short segments, or peptides, within this stable shell that the human immune system can recognize, they might be able to train the body to attack the virus regardless of how much it tries to disguise itself.

To find these hidden targets, a researcher turned to the power of computers to sift through a vast library of viral data. Instead of testing thousands of samples in a lab, which would be slow and expensive, they built a digital pipeline to scan 185 different sequences of the HIV-1 p24 protein collected from public databases. They lined these sequences up side by side, like comparing the pages of many different editions of the same book to find the words that never change. Using a method that measures how much variety exists at each position, they identified the spots where the virus is most consistent. From this analysis, they extracted thousands of tiny overlapping fragments, filtering them down to find the ones that appear most frequently across the different viral strains.

The result of this massive digital search was a ranked list of twenty-one specific peptide candidates that stand out for their stability and presence. The top contender, a nine-unit chain of amino acids known as GPKEPFRDY, appeared in nearly every single sequence the team examined, showing up in 182 out of 185 samples. This fragment, which is the right size to be presented to the immune system's T-cells, achieved a final score of 0.590 in their system, earning it the highest priority. The researcher also identified a longer fragment, KVVEEKAFSPEVIPM, which was supported by previous studies suggesting it triggers a different type of immune response involving antibodies. While the computer model did not test whether these fragments actually work in a living human, the sheer consistency of their presence across the viral population suggests they are strong candidates for further study.

This work does not claim to have solved the vaccine problem or proven that these peptides will stop the virus. The study explicitly stops short of testing how well these fragments bind to human immune cells or whether they can protect a person from infection. Instead, the researcher has provided a clear, reproducible map for other scientists to follow. By making their code and data public, they have handed the scientific community a shortlist of the most promising targets, saving time and resources that would otherwise be spent guessing which parts of the virus to investigate. The path forward now lies in the laboratory, where these computer-selected candidates must be tested to see if they can truly become the foundation of a new generation of HIV vaccines.

Technical Summary: Computational Identification and Prioritization of Conserved HIV-1 p24 Peptide Candidates

Problem Statement
Despite decades of research, an effective prophylactic vaccine for HIV-1 remains elusive, primarily due to the virus's extraordinary genetic diversity, particularly in the envelope glycoprotein (Env). While the p24 capsid protein is a relatively conserved structural domain essential for viral assembly and uncoating, and has been linked to CD8+ T-cell mediated viral control, there is a need for systematic methods to identify and prioritize specific peptide candidates within p24 for experimental vaccine development. The challenge lies in navigating the vast sequence variation of HIV-1 to isolate conserved regions that are both immunologically relevant and broadly representative of circulating strains.

Methodology
The author developed a fully reproducible, open-source computational pipeline to screen, analyze, and rank peptide candidates within the HIV-1 p24 protein. The workflow involved the following steps:

Data Acquisition and Quality Control: 185 HIV-1 p24 sequences (including partial coding sequences) were retrieved from the NCBI GenBank database. Sequences were filtered to ensure lengths between 200–260 amino acids, excluded those with >5% ambiguous residues, and deduplicated.
Multiple Sequence Alignment (MSA): The filtered dataset was aligned using MAFFT (v7.490) with iterative refinement, resulting in a 235-position alignment.
Conservation Analysis: Position-wise conservation was quantified using Shannon entropy. Conservation scores were calculated as $1 - (H(i) / \log_2(20))$ , where positions with scores >0.7 were deemed highly conserved.
Peptide Screening: Overlapping 9-mer (MHC-I length) and 15-mer (MHC-II length) peptides were extracted. The study explicitly did not perform direct HLA binding predictions (e.g., NetMHCpan); instead, it used peptide length as a structural proxy for potential MHC presentation.
Evidence Integration: Candidates were evaluated based on three criteria:
- Sequence Conservation: Average conservation score of the peptide region.
- Sequence Support: The frequency of the peptide's occurrence across the 185 sequences.
- B-cell Evidence: A binary flag indicating whether the peptide was supported by literature or databases (Los Alamos HIV Database, IEDB).
Ranking Framework: A weighted scoring system (40% Conservation, 30% B-cell Evidence, 30% Sequence Support) was applied to generate a final prioritized list.

Key Results

Dataset Characteristics: The analysis of 185 unique sequences revealed that the p24 protein is broadly conserved, with an average conservation score of 0.946. 95% of alignment positions (223/235) exceeded the 0.7 conservation threshold.
Peptide Identification: From the dataset, 10,112 unique peptides were identified (3,557 9-mers and 6,555 15-mers).
Top Candidates: The multi-criteria scoring system prioritized 21 candidates.
- The top-ranked candidate was the MHC-I length peptide GPKEPFRDY, which appeared in 182 of 185 sequences (98.4%) with a conservation score of 0.976 and a final weighted score of 0.590.
- The highest-ranked B-cell-supported candidate was the MHC-II length peptide KVVEEKAFSPEVIPM, supported by 41 sequences (22.2%) with a conservation score of 0.974 and a final score of 0.480.
Correlations: A positive relationship was observed between conservation scores and sequence support, indicating that highly conserved regions tend to be present across a larger proportion of the dataset.

Key Contributions

Reproducible Pipeline: The study provides a transparent, open-source computational framework (Python-based) for epitope prioritization that integrates sequence retrieval, alignment, conservation analysis, and multi-criteria ranking.
Prioritized Candidate List: A ranked list of 21 conserved HIV-1 p24 peptide candidates is provided for downstream experimental investigation.
Public Resource: All code, processed data, intermediate tables, and results are made publicly available to facilitate reproducibility and further research.

Significance and Claims
The author positions this work as a computational prioritization framework rather than a study of experimental validation. The significance of the paper lies in its ability to translate large-scale sequence data into a focused, data-driven list of candidates, which is particularly valuable for resource-limited settings where high-throughput experimental screening is not feasible.

The paper explicitly states that it does not claim to have validated immunogenicity, HLA binding affinity, or protective efficacy. Instead, it offers a ranked list of candidates to inform future experimental validation efforts, such as T-cell response assays (ELISPOT, MHC multimer) and B-cell epitope validation. The author emphasizes that the identification of these candidates is a necessary first step to guide the selection of targets for the development of T-cell-based vaccine strategies against HIV-1.

Technical Summary: Computational Identification and Prioritization of Conserved HIV-1 p24 Peptide Candidates

More like this