pith. sign in

arxiv: 2606.02462 · v2 · pith:HZCJC5UAnew · submitted 2026-06-01 · 🧬 q-bio.BM

APLSuite: An Integrated Suite for CD4+ T Cell Epitope Prediction via Antigen Processing Likelihood

Jiarui Li , Marco K. Carbullido , Jai Bansal , Samuel J. Landry , Ramgopal R. Mettu This is my paper

Pith reviewed 2026-06-28 11:15 UTC · model grok-4.3

classification 🧬 q-bio.BM
keywords epitope predictionCD4+ T cellsantigen processingsoftware suitecomputational immunologyAPL algorithmGPU acceleration
0
0 comments X

The pith

APLSuite supplies an integrated pipeline for CD4+ T cell epitope prediction that runs APL calculations in minutes with GPU support.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents APLSuite as a lightweight software suite that combines RESTful API services, a Python client, data science computation tools, and a graphical interface. Its purpose is to make the Antigen Processing Likelihood algorithm practical for everyday use in epitope prediction. The algorithm itself folds in crystallographic B-factor, solvent accessible surface area, hydrogen exchange protection factors, and sequence entropy to reflect how antigen processing shapes binding specificity. If the suite works as described, researchers gain a complete workflow deployable on desktop or cloud systems that finishes predictions faster than earlier standalone methods. This matters for immunology work because it turns a specialized calculation into a routine, accessible step in studying adaptive immune responses.

Core claim

APLSuite integrates distributed RESTful API services, a Python client for data aggregation and processing, a data science tool for efficient epitope computation, and a user-friendly graphical user interface. It supplies a seamless pipeline for APL calculation and epitope prediction that finishes in minutes under GPU acceleration, a capability previous tools did not provide. The suite supports both guided and customizable workflows and runs on desktop or cloud environments.

What carries the argument

APLSuite, a software suite that bundles API services, Python client, data science tools, and GUI to execute the APL algorithm for epitope prediction.

If this is right

  • Epitope prediction becomes feasible as a routine step completed in minutes rather than hours or days.
  • Non-programming researchers gain access through the graphical interface without writing code.
  • The same pipeline supports both quick guided runs and custom parameter adjustments.
  • Deployment on either local desktops or cloud servers removes hardware barriers for different labs.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the processing factors prove reliable, the same modular structure could be adapted to predict epitopes for other T cell types or MHC classes.
  • Large-scale screening of pathogen proteomes becomes practical once the minutes-scale runtime is available.
  • Integration with existing sequence databases could allow automated updates to input structures without manual downloads.

Load-bearing premise

The earlier APL algorithm, which combines B-factor, SASA, COREX, and sequence entropy, actually captures how antigen processing influences CD4+ T cell epitope binding.

What would settle it

A direct comparison in which APL-based predictions show no better match to measured CD4+ T cell responses than predictions made without the processing factors.

Figures

Figures reproduced from arXiv: 2606.02462 by Jai Bansal, Jiarui Li, Marco K. Carbullido, Ramgopal R. Mettu, Samuel J. Landry.

Figure 1
Figure 1. Figure 1: APLSuite Framework Overview: The APLSuite framework including Distributed RESTful API (DRAF), Python client, data science tool, and graphical user interface (GUI). The APL components and itself are developed as API endpoints, remote Python function, and web UI based tool using this framework. It is network service based on Django, FastAPI, and Docker. The storage engine supporting SQLite, MongoDB, and Mong… view at source ↗
Figure 2
Figure 2. Figure 2: Graphical User Interface (GUI): The views of GUI including project view, tool view, quick start view, and resource management view. Which enables both step-by-step guided running and highly flexible workbench. It also supports to download the results as a report or a compressed file of CSV tables. The project view ( [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Distributed RESTful API Framework (DRAF): [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Data Science Tool (DST): Data Science Tool (DST) streamlines access to DRAF’s functionality and make it more accessible to a wide range of users • Algorithm Chaining: Leveraging DRAF’s standardized data type protocol, the DST allows users to chain multiple algorithms by providing a sequence of API endpoint indices. This simplifies the creation of complex workflows. In addition, for user customized chained … view at source ↗
Figure 5
Figure 5. Figure 5: Application of the APL-MHC pipeline the VSV gE antigen. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Application of the APL-MHC pipeline to a set of antigens. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
read the original abstract

Computational epitope prediction is a critical tool for exploring and understanding CD4+ T cell-mediated immune responses, a key aspect of adaptive immunity. While existing computational methods primarily focus on supervised learning approaches, they often overlook the essential role of antigen processing in determining binding specificity. To address this limitation, our group developed Antigen Processing Likelihood (APL), an algorithm that integrates crystallographic B-factor, solvent accessible surface area (SASA), hydrogen exchange protection factors (COREX), and sequence entropy. In this paper we introduce APLSuite, a comprehensive and lightweight software suite designed to streamline APL-based epitope prediction. APLSuite integrates distributed RESTful API services, a Python client for data aggregation and processing, a data science tool for efficient epitope computation, and a user-friendly graphical user interface for non-coding users. It provides a seamless and efficient pipeline for APL calculation and epitope prediction that can be finished in minutes with GPU-acceleration, which has not been implemented by existed tools. This flexible and extensible software suite is deployable on desktop and cloud environments, offering both guided and customizable workflows to meet diverse research needs in immunology research and immunotherapy development. (The project page for this work is available at: https://tulane-mettu-landry-lab.github.io/blogs/APLSuite/)

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces APLSuite, a software suite for CD4+ T cell epitope prediction based on the Antigen Processing Likelihood (APL) algorithm. APL integrates B-factor, SASA, COREX, and sequence entropy. The suite comprises distributed RESTful API services, a Python client for data aggregation, a data science tool for epitope computation, and a GUI. It claims to deliver a seamless pipeline for APL calculation and epitope prediction that completes in minutes with GPU acceleration and has not been implemented by existing tools. The software is described as flexible, extensible, and deployable on desktop or cloud environments.

Significance. If the efficiency and uniqueness claims hold and are substantiated with benchmarks, APLSuite would provide a practical, accessible implementation of an antigen-processing-aware epitope prediction method. This could support immunology research and immunotherapy development by lowering barriers for both coding and non-coding users while addressing limitations of purely supervised approaches.

major comments (2)
  1. [Abstract] Abstract: The claim that the pipeline 'can be finished in minutes with GPU-acceleration, which has not been implemented by existed tools' is load-bearing for the central contribution yet is unsupported by any runtime data, GPU-vs-CPU speedup measurements, scaling curves, or head-to-head comparisons against prior epitope-prediction packages.
  2. [Abstract] Abstract: No validation data, benchmark comparisons against known CD4+ T cell epitopes, error analysis, or experimental confirmation of prediction accuracy are supplied, leaving the practical utility of the APL integration unevaluable.
minor comments (2)
  1. [Abstract] Abstract: Typo 'existed tools' should read 'existing tools'.
  2. [Abstract] Abstract: The phrase 'non-coding users' is unclear; consider 'users without programming experience'.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments on the APLSuite manuscript. We address each major comment below and commit to revisions where the concerns are valid.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that the pipeline 'can be finished in minutes with GPU-acceleration, which has not been implemented by existed tools' is load-bearing for the central contribution yet is unsupported by any runtime data, GPU-vs-CPU speedup measurements, scaling curves, or head-to-head comparisons against prior epitope-prediction packages.

    Authors: We agree this claim lacks supporting data in the current manuscript. In revision we will add a Performance Evaluation section with runtime benchmarks (CPU vs. GPU), scaling behavior on sample antigens, and limited comparisons to existing packages such as NetMHCIIpan. If generating full head-to-head data exceeds revision timeline, we will qualify or remove the timing claim from the abstract while preserving the description of the integrated APL pipeline. revision: yes

  2. Referee: [Abstract] Abstract: No validation data, benchmark comparisons against known CD4+ T cell epitopes, error analysis, or experimental confirmation of prediction accuracy are supplied, leaving the practical utility of the APL integration unevaluable.

    Authors: The manuscript is a software description paper whose scope is the implementation of the APLSuite components rather than new validation of the APL algorithm (previously reported by our group). To address utility concerns we will add a brief case-study subsection in revision that applies the suite to example antigens and notes alignment with publicly documented CD4+ epitopes, thereby illustrating output without new experimental data. revision: yes

Circularity Check

0 steps flagged

No circularity: software implementation paper with no derivation or fitted predictions

full rationale

The manuscript describes an engineering integration (REST APIs, Python client, GUI, GPU acceleration) around a previously published APL algorithm developed by the same group. No equations, parameter fitting, predictions, or uniqueness theorems appear in the provided text. The central claim is that the suite delivers a pipeline 'not implemented by existed tools,' but this is an empirical assertion about software availability rather than a mathematical reduction of any result to its own inputs. Self-citation of the prior APL work is present but not load-bearing for any derivation within this paper; the work is self-contained as an implementation layer and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The abstract introduces no free parameters, mathematical axioms, or new postulated entities; the work is a software engineering description of an existing computational method.

pith-pipeline@v0.9.1-grok · 5778 in / 1272 out tokens · 36369 ms · 2026-06-28T11:15:52.666627+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

25 extracted references

  1. [1]

    Human T cell development, localization, and function throughout life

    Kumar BV, Connors TJ, Farber DL. Human T cell development, localization, and function throughout life. Immunity. 2018;48(2):202–213

    2018

  2. [2]

    Accurate prediction of HLA class II antigen presentation across all loci using tailored data acquisition and refined machine learning

    Nilsson JB, Kaabinejadian S, Yari H, Kester MG, van Balen P, Hildebrand WH, et al. Accurate prediction of HLA class II antigen presentation across all loci using tailored data acquisition and refined machine learning. Science Advances. 2023;9(47):eadj6367. 8

    2023

  3. [3]

    Machine learning reveals limited contribution of trans-only encoded variants to the HLA-DQ immunopeptidome

    Nilsson JB, Kaabinejadian S, Yari H, Peters B, Barra C, Gragert L, et al. Machine learning reveals limited contribution of trans-only encoded variants to the HLA-DQ immunopeptidome. Communications biology. 2023;6(1):442

    2023

  4. [4]

    The immune epitope database (IEDB): 2018 update

    Vita R, Mahajan S, Overton JA, Dhanda SK, Martini S, Cantrell JR, et al. The immune epitope database (IEDB): 2018 update. Nucleic acids research. 2019;47(D1):D339–D343

    2018

  5. [5]

    iedb.org: free epitope database and prediction resource; 2024

    IEDB. iedb.org: free epitope database and prediction resource; 2024. Available from: http://www. iedb.org/

    2024

  6. [6]

    CD4+ T-cell epitope prediction using antigen processing constraints

    Mettu RR, Charles T, Landry SJ. CD4+ T-cell epitope prediction using antigen processing constraints. Journal of immunological methods. 2016;432:72–81

    2016

  7. [7]

    Cd4+ T-cell epitope prediction by combined analysis of antigen conformational flexibility and peptide-Mhcii binding affinity

    Charles T, Moss DL, Bhat P, Moore PW, Kummer NA, Bhattacharya A, et al. Cd4+ T-cell epitope prediction by combined analysis of antigen conformational flexibility and peptide-Mhcii binding affinity. Biochemistry. 2022;61(15):1585–1599

    2022

  8. [8]

    Structural Framework for Analysis of CD4+ T-Cell Epitope Dominance in Viral Fusion Proteins

    Landry SJ, Mettu RR, Kolls JK, Aberle JH, Norton E, Zwezdaryk K, et al. Structural Framework for Analysis of CD4+ T-Cell Epitope Dominance in Viral Fusion Proteins. Biochemistry. 2023;62(17):2517– 2529

    2023

  9. [9]

    Structure-based calculation of the equilibrium folding pathway of proteins

    Hilser VJ, Freire E. Structure-based calculation of the equilibrium folding pathway of proteins. Correlation with hydrogen exchange protection factors. Journal of molecular biology. 1996;262(5):756– 772

    1996

  10. [10]

    FreeSASA: An open source C library for solvent accessible surface area calculations

    Mitternacht S. FreeSASA: An open source C library for solvent accessible surface area calculations. F1000Research. 2016;5

    2016

  11. [11]

    COREX/BEST server: a web browser-based program that calculates regional stability variations within protein structures

    Vertrees J, Barritt P, Whitten S, Hilser VJ. COREX/BEST server: a web browser-based program that calculates regional stability variations within protein structures. Bioinformatics. 2005;21(15):3318–3319

    2005

  12. [12]

    Parallel Computation of Conformational Stability for CD4+ T-cell Epitope Prediction

    Bhattacharya A, Wrabl JO, Landry SJ, Mettu RR. Parallel Computation of Conformational Stability for CD4+ T-cell Epitope Prediction. In: 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE; 2023. p. 88–93

    2023

  13. [13]

    GPU Acceleration of Conformational Stability Computation for CD4+ T-cell Epitope Prediction

    Li J, Landry SJ, Mettu RR. GPU Acceleration of Conformational Stability Computation for CD4+ T-cell Epitope Prediction. In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE; 2024. p. 191–196

    2024

  14. [14]

    GPU Acceleration for Markov Chain Monte Carlo Sampling

    Li J, Landry SJ, Mettu RR. GPU Acceleration for Markov Chain Monte Carlo Sampling. In: 4th International Conference on AIML Systems (AIMLSystems 2024). ACM; 2024. p. 1–8

    2024

  15. [15]

    Basic local alignment search tool

    Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. Journal of molecular biology. 1990;215(3):403–410

    1990

  16. [16]

    BLAST+: architecture and applications

    Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC bioinformatics. 2009;10:1–9

    2009

  17. [17]

    Multiple sequence alignment

    Edgar RC, Batzoglou S. Multiple sequence alignment. Current opinion in structural biology. 2006;16(3):368–373

    2006

  18. [18]

    Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega

    Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Molecular systems biology. 2011;7(1):539

    2011

  19. [19]

    Clustal Omega for making accurate alignments of many protein sequences

    Sievers F, Higgins DG. Clustal Omega for making accurate alignments of many protein sequences. Protein Science. 2018;27(1):135–145

    2018

  20. [20]

    The web framework for perfectionists with deadlines.; 2024

    Django. The web framework for perfectionists with deadlines.; 2024. Available from: https://www. djangoproject.com/. 9

    2024

  21. [21]

    FastAPI framework, high performance, easy to learn, fast to code, ready for production; 2024

    FastAPI. FastAPI framework, high performance, easy to learn, fast to code, ready for production; 2024. Available from:https://fastapi.tiangolo.com/

    2024

  22. [22]

    Deciphering and predicting CD4+ T cell immunodominance of influenza virus hemagglutinin

    Cassotta A, Paparoditis P, Geiger R, Mettu RR, Landry SJ, Donati A, et al. Deciphering and predicting CD4+ T cell immunodominance of influenza virus hemagglutinin. Journal of Experimental Medicine. 2020;217(10):e20200206

    2020

  23. [23]

    Identification and characterization of CD4+ T cell epitopes after Shingrix vaccination

    Voic H, de Vries RD, Sidney J, Rubiro P, Moore E, Phillips E, et al. Identification and characterization of CD4+ T cell epitopes after Shingrix vaccination. Journal of Virology. 2020;94(24):10–1128

    2020

  24. [24]

    Development and validation of a broad scheme for prediction of HLA class II restricted T cell epitopes

    Paul S, Sidney J, Peters B, Sette A. Development and validation of a broad scheme for prediction of HLA class II restricted T cell epitopes. In: Proceedings of the 5th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics; 2014. p. 733–738

    2014

  25. [25]

    Incorporating antigen processing into CD4+ T cell epitope prediction with integer linear programming

    Bhattacharya A, Lyons MC, Landry SJ, Mettu RR. Incorporating antigen processing into CD4+ T cell epitope prediction with integer linear programming. In: Proceedings of the 13th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics; 2022. p. 1–10. 10

    2022