YLYifeng Li

Yifeng Li

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Yifeng Li has a total of 97 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Z-Hunt-DP: accelerating thermodynamic Z-DNA prediction with dynamic programming

    by Maysara Al Jumaily, Hamza Qureshi, Hongbin Yan, Yifeng Li

    Z-DNA is a left-handed DNA conformation implicated in gene regulation and chromatin dynamics. Because it is usually less thermodynamically favorable than canonical B-DNA under physiological conditions, computational tools are needed to identify sequences likely to adopt the Z conformation. Legacy Z-Hunt uses a dinucleotide thermodynamic model, but searches every anti/syn assignment in a window, causing its conformation search to grow exponentially with window size. We present Z-Hunt-DP, an exact dynamic programming reformulation that preserves the original thermodynamic objective while reducing this search from O (2 d ) to O ( d ) for a window of d dinucleotide positions. On benchmark windows, Z-Hunt-DP matched the brute-force minimum energy within numerical tolerance and achieved a 3.30 × 10 5 speedup at 20 dinucleotides. In an interval-localization benchmark on public human loci with experimentally mapped Z-DNA, it recovered the clipped reference interval in all 24 cases and was the most stable localizer under midpoint-core and expanded-panel analyses. Since the comparison set mixes thermodynamic, heuristic, and learned models, these cross-tool results are interpreted as localization comparisons rather than direct thermodynamic score tests. The source code and benchmark materials are available at https://github.com/Aljumaily/Z-Hunt-DP . Author summary: DNA can form structures other than its familiar right-handed double helix. One alternative, Z-DNA, is left-handed and may influence how genes are regulated, but identifying it across long DNA sequences remains computationally demanding. Existing Z-Hunt software estimates Z-DNA propensity from experimentally derived energy values. Its central calculation tests every possible local DNA conformation, so its run time rises very quickly as the sequence window grows. We developed Z-Hunt-DP, which solves the same thermodynamic problem by reusing the best partial solutions at each position. This yields the same optimal energy as exhaustive search but requires far less computation. We confirmed exact agreement with brute-force calculations and found large speed gains on increasing window sizes. We also tested whether predicted regions align with public human loci where Z-DNA has been experimentally mapped. Z-Hunt-DP consistently located the center of those loci more reliably than the other tested tools. The open-source implementation makes this established thermodynamic model practical for reproducible genome-scale analysis and for prioritizing compact experimental validation targets.

  • Open Access

    Reply to “Letter to the editor: methodological considerations in the benchmarking of AI-based protein–aptamer complex prediction”

    by Jiani Zhao, Kha Tram, Hongbin Yan, Yifeng Li

    We thank MD Kazim Okan Dolu for their interest in our published work and for the opportunity to clarify several points regarding software versioning, benchmark independence, secondary-structure generation, binding-free-energy terminology, and negative-control design. We acknowledge that several descriptions in the original manuscript could be stated more precisely. These clarifications improve the transparency and reproducibility of the study, but do not affect the reported analyses or main conclusions.

  • Open Access

    Comprehensive evaluation of artificial intelligence-empowered approaches for protein–aptamer complex prediction

    by Jiani Zhao, Kha Tram, Hongbin Yan, Yifeng Li

    Drug discovery is a time-consuming, expensive, and high-risk process. Recent advances in artificial intelligence (AI) have enabled major breakthroughs in small-molecule and protein therapeutics. However, AI-driven design of aptamer drugs remains largely unexplored. Aptamers are short (15–100 nt) single-stranded DNAs or RNAs that exhibit high binding affinity, high specificity, and low immunogenicity, making them promising candidates for disease (such as cancer) therapeutics. Compared with protein–ligand or protein–protein systems, protein–aptamer complexes are under-represented in public structural databases, and aptamers themselves are highly flexible and relatively large molecules. These characteristics present distinct challenges for AI-based structural modeling. Here, we systematically evaluate recent AI frameworks, including AlphaFold3, Chai-1, Boltz-2, and RoseTTAFold2NA, along with a template-based approach, in predicting protein–aptamer complex structures and estimating binding free energies. We establish an independent benchmark to assess their performance in structural accuracy, stability, and energetic consistency. This study provides a foundation for the application of AI in aptamer drug design and offers a reference framework for future research in nucleic-acid therapeutics and biomolecular modeling.


  • Open Access

    Aligning Language Models Using Multi-Objective Deep Reinforcement Learning

    by Yage Zhang, Xiaodan Zhu, Yifeng Li

    The alignment techniques used in state-of-the-art language models (LMs), e.g., reinforcement learning from human feedback (RLHF), have driven many successful natural language processing (NLP) tasks. RLHF uses human preferences based on the guidelines of being helpful and safe as a single reward signal to fine-tune language models. However, the trade-offs between helpfulness and safety are often found to be a problem, which makes it difficult for a model trained towards one objective to perform well on both. This paper proposes a new alignment technique, multi-objective language model alignment (MOLMA). The framework is based on multi-objective deep reinforcement learning to fine-tune language models. MOLMA can efficiently address the conflicting or the dominating learning signal issue caused by the trade-offs of inherent, often conflicting, multi-objectives underlying the language model alignment task. From the overall objective of achieving helpfulness and safety, our results show that MOLMA outperforms the other alignment techniques that rely on single-objective deep reinforcement learning.


  • Open Access

    Exploration of chaos game representation and integrative deep learning approaches for whole-genome sequencing-based grapevine genetic testing

    by Andrew Vu, Brendan Park, Yifeng Li, Ping Liang

    Motivation: The identification of grapevine species, cultivars, and clones associated with desired traits is an important component of viticulture. True-to-type identification is very challenging for grapevine due to the existence of a large number of cultivars and clones and the historical issues of synonyms and homonyms. DNA-based identification, superior to morphology-based methods, has been used as the current standard true-to-type method for grapevine, but not without shortcomings, such as the limited number of biomarkers and accessibility of services. Results: To overcome some of the limitations of traditional microsatellite-marker-based genetic testing, we explored a whole-genome-sequencing (WGS)-based approach to achieve the best accuracy at an affordable cost. To address the challenges of the extreme high dimensionality of the WGS data, we examined the effectiveness of using chaos game representation (CGR) to represent the genome sequence data and using deep learning for species and cultivar identification. CGR images provide a meaningful way to capture patterns for use with visual analysis, with the best results showing a 99% balanced accuracy in classifying five species, and a 80% balanced accuracy in predicting 41 cultivars. Our preliminary research highlights the potential for CGR and deep learning as a complementary tool for WGS-based species- and cultivar-level classification. Availability and implementation: Our implementation, including the pipeline for data processing and the four predictive models, is available at https://github.com/pliang64/CGR.


  • “Several birds with one stone”: exploring the potential of AI methods for multi-target drug design

    by Muhetaer Mukaidaisi, Madiha Ahmed, Karl Grantham, Aws Al-Jumaily, Shoukat Dedhar, Michael Organ, Alain Tchagang, Jinqiang Hou, Syed Ejaz Ahmed, Renata Dividino, et al.

    Drug discovery is a time-consuming and expensive process. Artificial intelligence (AI) methodologies have been adopted to cut costs and speed up the drug development process, serving as promising in silico approaches to efficiently design novel drug candidates targeting various health conditions. Most existing AI-driven drug discovery studies follow a single-target approach which focuses on identifying compounds that bind a target (i.e., one-drug-one-target approach). Polypharmacology is a relatively new concept that takes a systematic approach to search for a compound (or a combination of compounds) that can bind two or more carefully selected protein biomarkers simultaneously to synergistically treat the disease. Recent studies have demonstrated that multi-target drugs offer superior therapeutic potentials compared to single-target drugs. However, it is intuitively thought that searching for multi-target drugs is more challenging than finding single-target drugs. At present, it is unclear how AI approaches perform in designing multi-target drugs. In this paper, we comprehensively investigated the performance of multi-objective AI approaches for multi-target drug design. Our findings are quite counter-intuitive demonstrating that, in fact, AI approaches for multi-target drug design are able to efficiently generate more high-quality novel compounds than the single-target approaches while satisfying a number of constraints.


  • Open Access

    Hybrid fragment-SMILES tokenization for ADMET prediction in drug discovery

    by Nicholas Aksamit, Alain Tchagang, Yifeng Li, Beatrice Ombuki-Berman

    Background: Drug discovery and development is the extremely costly and time-consuming process of identifying new molecules that can interact with a biomarker target to interrupt the disease pathway of interest. In addition to binding the target, a drug candidate needs to satisfy multiple properties affecting absorption, distribution, metabolism, excretion, and toxicity (ADMET). Artificial intelligence approaches provide an opportunity to improve each step of the drug discovery and development process, in which the first question faced by us is how a molecule can be informatively represented such that the in-silico solutions are optimized. Results: This study introduces a novel hybrid SMILES-fragment tokenization method, coupled with two pre-training strategies, utilizing a Transformer-based model. We investigate the efficacy of hybrid tokenization in improving the performance of ADMET prediction tasks. Our approach leverages MTL-BERT, an encoder-only Transformer model that achieves state-of-the-art ADMET predictions, and contrasts the standard SMILES tokenization with our hybrid method across a spectrum of fragment library cutoffs. Conclusion: The findings reveal that while an excess of fragments can impede performance, using hybrid tokenization with high frequency fragments enhances results beyond the base SMILES tokenization. This advancement underscores the potential of integrating fragment- and character-level molecular features within the training of Transformer models for ADMET property prediction.


  • Open Access

    Integrating transformers and many-objective optimization for drug design

    by Nicholas Aksamit, Jinqiang Hou, Yifeng Li, Beatrice Ombuki-Berman

    Background: Drug design is a challenging and important task that requires the generation of novel and effective molecules that can bind to specific protein targets. Artificial intelligence algorithms have recently showed promising potential to expedite the drug design process. However, existing methods adopt multi-objective approaches which limits the number of objectives. Results: In this paper, we expand this thread of research from the many-objective perspective, by proposing a novel framework that integrates a latent Transformer-based model for molecular generation, with a drug design system that incorporates absorption, distribution, metabolism, excretion, and toxicity prediction, molecular docking, and many-objective metaheuristics. We compared the performance of two latent Transformer models (ReLSO and FragNet) on a molecular generation task and show that ReLSO outperforms FragNet in terms of reconstruction and latent space organization. We then explored six different many-objective metaheuristics based on evolutionary algorithms and particle swarm optimization on a drug design task involving potential drug candidates to human lysophosphatidic acid receptor 1, a cancer-related protein target. Conclusion: We show that multi-objective evolutionary algorithm based on dominance and decomposition performs the best in terms of finding molecules that satisfy many objectives, such as high binding affinity and low toxicity, and high drug-likeness. Our framework demonstrates the potential of combining Transformers and many-objective computational intelligence for drug design.


  • Open Access

    "Several Birds with One Stone": Exploring the Potential of AI Methods for Multi-Target Drug Design

    by Muhetaer Mukaidaisi, Madiha Ahmed, Karl Grantham, Aws Al-Jumaily, Shoukat Dedhar, Michael Organ, Alain Tchagang, Jinqiang Hou, Ejaz Ahmed, Renata Dividino, et al.

    Background: Drug discovery is a time-consuming and expensive process. Artificial intelligence (AI) methodologies have been adopted to cut costs and speed up the drug development process, serving as promising in silico approaches to efficiently design novel drug candidates targeting various health conditions. Most existing AI-driven drug discovery studies follow a single-target approach which focuses on identifying compounds that bind a single target (i.e., one-drug-one-target approach). Polypharmacology is a relatively new concept that takes a systematic approach to search for a compound (or a combination of compounds) that can bind two or more carefully selected protein biomarkers simultaneously to synergistically treat the disease. Recent studies have demonstrated that multi-target drugs offer superior therapeutic potentials compared to single-target drugs. However, it is intuitively thought that searching for multi-target drugs is more challenging than finding single-target drugs. At present, it is unclear how AI approaches perform in designing multi-target drugs. Results: In this paper, we comprehensively investigated the performance of multi-objective AI approaches for multi-target drug design. Conclusion: Our findings are quite counterintuitive demonstrating that, in fact, AI approaches for multi-target drug design are able to efficiently generate more high-quality novel compounds than the single-target approaches while satisfying a number of constraints.

  • Open Access

    Integrating Transformers and Many-Objective Optimization for Cancer Drug Design

    by Nicholas Aksamit, Jinqiang Hou, Yifeng Li, Beatrice Ombuki-Berman

    Background: Drug design is a challenging and important task that requires the generation of novel and effective molecules that can bind to specific protein targets. Artificial intelligence (AI) algorithms have recently showed promising potential to expedite the drug design process. However, existing methods adopt multi-objective approaches which limits the number of objectives. Results: In this paper, we expand this thread of research from the many-objective perspective, by proposing a novel framework that integrates a latent Transformer-based model for molecular generation, with a drug design system that incorporates absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction, molecular docking, and many-objective metaheuristics. We compared the performance of two latent Transformer models (ReLSO and FragNet) on a molecular generation task and show that ReLSO outperforms FragNet in terms of reconstruction and latent space organization. We then explored six different many-objective metaheuristics based on evolutionary algorithms and particle swarm optimization on a drug design task involving potential drug candidates to human lysophosphatidic acid receptor 1 (LPA1), a cancer-related protein target. Conclusion: We show that multi-objective evolutionary algorithm based on dominance and decomposition (MOEA/DD) performs the best in terms of finding molecules that satisfy many objectives, such as high binding affinity and low toxicity, and high drug-likeness. Our framework demonstrates the potential of combining Transformers and many-objective computational intelligence for cancer drug design.