AI, CRISPR and Multi-Omics Converge to Reshape Antileishmanial Drug Discovery

Leishmaniasis, a parasitic disease transmitted by sand flies, continues to burden millions of people across tropical and subtropical regions, yet its drug pipeline remains strikingly thin. A new review published in Molecular Biology Reports by Derya Topuz Ata, Anıl Ata and Zeynep Tuba Odacı of Ankara University argues that the path to better therapies lies not in any single technology, but in the deliberate integration of multi-omics profiling, CRISPR-based functional genomics, artificial intelligence and host-directed immunotherapy into one coherent discovery framework. The authors contend that only such a combined, precision-oriented strategy can produce antileishmanial drugs that are safer, faster-acting and resilient against the parasite’s notorious ability to evolve resistance.

The scale of the problem justifies the ambition. Current treatments for leishmaniasis, including pentavalent antimonials, amphotericin B, miltefosine, paromomycin and sitamaquine, are constrained by toxicity, high cost, long treatment courses, treatment failure and the steady emergence of drug-resistant parasites. These shortcomings are particularly damaging for visceral leishmaniasis, the most dangerous form of the disease, and for cutaneous and mucosal manifestations that leave lasting scars and stigma. Because no licensed human vaccine exists, chemotherapy remains the principal line of defence, which makes the erosion of drug efficacy a direct threat to public health in endemic countries.

Compounding the therapeutic challenge is the peculiar biology of Leishmania itself. The parasites display remarkable genomic plasticity, characterised by mosaic aneuploidy, in which individual chromosomes exist in variable copy numbers within the same population, alongside widespread copy number variation. This fluid genome architecture allows rapid adaptation under drug pressure and complicates the identification of durable molecular targets, since a gene that appears essential in one strain or life stage may be buffered by extra copies in another. The review emphasises that any credible target-discovery effort must therefore account for this adaptive capacity, favouring approaches that validate essentiality and resistance liability experimentally rather than relying on static genome annotations alone.

Multi-omics technologies form the first pillar of the proposed framework. Comparative genomics across Leishmania species, transcriptome diversity mapped by nanopore direct RNA sequencing, proteomic surveys of organelles such as glycosomes, lipidomic analyses of extracellular vesicles from resistant parasites, and single-cell transcriptomics of infected macrophages together reveal the pathways that sustain parasite survival, virulence and drug tolerance. Recent single-cell studies have exposed rare parasitised cell populations in chronic Leishmania donovani infection and lipid metabolism reprogramming in macrophages during early infection, while spatial mapping of cutaneous lesions has uncovered distinct tissue-level immune programmes. These datasets collectively illuminate host–parasite interplay at unprecedented resolution and nominate candidate vulnerabilities for therapeutic intervention.

The second pillar is CRISPR-based genome engineering, which has matured from the first Cas9 edits in Leishmania in 2015 into a versatile toolbox for functional genomics. High-throughput CRISPR-Cas9 screens in Leishmania infantum have been used to dissect drug resistance, cytosine base editor toolboxes such as LeishBASEedit enable scalable loss-of-function studies, and CRISPR-Cas13b knockdowns extend editing to RNA-level perturbations. Systematic deletion screens, including the TransLeish effort that identified membrane transporters essential for intracellular survival, and resources such as LeishGEM, which catalogues genome-wide deletion mutant fitness and protein localisation, now allow researchers to confirm which candidate targets are genuinely indispensable. The review highlights how these tools also expose resistance mechanisms, from single amino acid substitutions conferring antimony tolerance to genes required for susceptibility to miltefosine.

Artificial intelligence constitutes the third pillar, converting the growing data flood into prioritised hypotheses. Machine-learning models trained on molecular fingerprints can predict leishmanial activity, ensemble techniques improve those predictions, and deep multitask learning has already guided the discovery of new compounds active against L. infantum. Graph-embedding approaches have uncovered potent candidate molecules, while AI-empowered platforms such as cidalsDB aggregate anti-pathogen therapeutics research. The authors describe how AI-assisted predictive modelling supports target prioritisation, virtual screening and structure-guided drug design, and they situate these methods within the broader rise of foundation models and digital twins in drug discovery, technologies that promise to compress timelines and reduce attrition when coupled with rigorous validation and uncertainty quantification.

From this convergence, the review identifies several high-priority therapeutic pathways. Redox homeostasis stands out because Leishmania relies on the trypanothione system, comprising trypanothione reductase and trypanothione synthetase, rather than the glutathione chemistry of its mammalian hosts, offering a selectivity window that inhibitor programmes are actively exploiting. Sterol biosynthesis provides another validated axis: the cytochrome P450 sterol 14-alpha-demethylase CYP51, its reductase partner P450R1, sterol C4-methyl oxidase and sterol C24-methyltransferase all influence membrane integrity and drug sensitivity. MAP kinase signalling, mitochondrial electron transport at the quinone reduction site of cytochrome b, and arginine metabolism round out the target landscape, each supported by genetic or pharmacological evidence of importance to parasite viability or infectivity.

The fourth pillar shifts the therapeutic lens from the parasite to the host. Host-directed immunotherapy seeks to reinvigorate the macrophage and T-cell responses that control infection, for example by promoting interferon-gamma signalling, inducible nitric oxide synthase activity and nitric oxide production through pathways involving Toll-like receptors, the JAK–STAT axis and mTOR regulation. Because such approaches do not exert direct pressure on parasite genes, they are inherently less likely to select for resistance and can complement parasite-directed drugs, potentially shortening regimens and improving outcomes in immunocompromised patients. The review surveys candidate host-directed agents and argues that combining them with precision antileishmanials represents a strategically sound route to resistance-resilient therapy.

The authors are candid about the obstacles ahead. Leishmania’s unusual gene expression regime, in which genes are transcribed in long polycistronic clusters and controlled largely post-transcriptionally, complicates both functional screens and AI models trained on conventional eukaryotic assumptions. Genetic diversity across species and within emerging transmission foci means that targets validated in one parasite may not translate to another, and data quality, model interpretability and experimental validation remain bottlenecks for computational discovery. Delivering host-directed therapies safely also demands careful immunological profiling to avoid exacerbating pathology in tissues where an overexuberant inflammatory response drives disease.

Nevertheless, the review’s central message is one of cautious optimism: the pieces of a precision drug-discovery pipeline for leishmaniasis now exist and can be connected. Omics surveys nominate pathways, CRISPR screens validate essentiality and map resistance, AI prioritises targets and designs molecules against them, and host-directed immunotherapy adds a complementary, resistance-sparing modality. If the research community and funders invest in integrating these strands, the authors conclude, the field can move beyond the incremental repurposing that has dominated recent decades and toward safer, more efficient and durable treatments for one of the world’s most neglected tropical diseases.

Subject of Research: Precision drug discovery for leishmaniasis using multi-omics, CRISPR functional genomics, artificial intelligence and host-directed therapeutics

Article Title: Towards precision antileishmanial drug discovery: Integrating multi-omics, functional genomics, artificial intelligence and host-directed therapeutics

Article References: TOPUZ ATA, D., ATA, A., & ODACI, Z. T. (2026). Towards precision antileishmanial drug discovery: Integrating multi-omics, functional genomics, artificial intelligence and host-directed therapeutics. Molecular Biology Reports, 53(1), Article 1620. https://doi.org/10.1007/s11033-026-12801-y

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12801-y

Keywords: leishmaniasis, multi-omics, CRISPR-Cas9, functional genomics, artificial intelligence, host-directed immunotherapy, drug resistance, trypanothione, sterol biosynthesis, neglected tropical diseases, drug discovery, precision medicine