Transfer learning is a method of learning new tasks in related domain using existing knowledge from source data. In rare disease research, data are often limited. Transfer learning can effectively use data from other related diseases or fields to enhance model performance and research efficiency. This approach helps researchers rapidly identify characteristics and develop potential treatments of rare disease. Currently, transfer learning has been applied in the systematic characterization and drug development of rare diseases. It also shows potential in optimizing rare disease classification, accelerating early diagnosis, and supporting multi-task research. However,  challenges arise in the application of transfer learning in rare disease research. In the future, if transfer learning can be combined with techniques such as reinforcement learning, federated learning, and deep learning, greater breakthroughs are expected to be achieved in the field of rare diseases.

1.周奇, 李沁原, 刘雅莉, 等. 罕见病指南的制订: 现状、挑战与机遇[J]. 协和医学杂志, 2023, 14(3): 621-628. [Zhou Q, Li QY, Liu YL, et al. The development of guidelines for rare diseases: past, present and future[J]. Medical Journal of Peking Union Medical College Hospital, 2023, 14(3): 621-628.] DOI: 10.12290/xhyxzz.2022-0360.

2.Denton N, Mulberg AE, Molloy M, et al. Sharing is caring: a call for a new era of rare disease research and development[J]. Orphanet J Rare Dis, 2022, 17(1): 389. DOI: 10.1186/s13023-022-02529-w.

3.Rees CA, Pica N, Monuteaux MC, et al. Noncompletion and nonpublication of trials studying rare diseases: a cross-sectional analysis[J]. PLoS Med, 2019, 16(11): e1002966. DOI: 10.1371/journal.pmed.1002966.

4.潘璐璐, 余勇夫, 秦国友. 迁移学习简介及其在医学研究领域中的应用[J]. 复旦学报（医学版）, 2024, 51(6): 1016-1020. [Pan LL, Yu YF, Qin GY. Introduction and application of transfer learning in medical research[J]. Fudan University Journal of Medical Sciences, 2024, 51(6): 1016-1020.] DOI: 10.3969/j.issn.1672-8467.2024.06.020.

5.Banerjee J, Taroni JN, Allaway RJ, et al. Machine learning in rare disease[J]. Nat Methods, 2023, 20(6): 803-814. DOI: 10.1038/s41592-023-01886-z.

6.Shyalika C, Wickramarachchi R, Sheth AP. A comprehensive survey on rare event prediction[J]. ACM Comput Surv, 2024, 57(3): 1-39. DOI: 10.1145/3699955.

7.Gao Y, Cui Y. Deep transfer learning for reducing health care disparities arising from biomedical data inequality[J]. Nat Commun, 2020, 11(1): 5131. DOI: 10.1038/s41467-020-18918-3.

8.Gao Y, Cui Y. Optimizing clinico-genomic disease prediction across ancestries: a machine learning strategy with Pareto improvement[J]. Genome Med, 2024, 16(1): 76. DOI: 10.1186/s13073-024-01345-0.

9.Taroni JN, Grayson PC, Hu Q, et al. MultiPLIER: a transfer learning framework for transcriptomics reveals systemic features of rare disease[J]. Cell Syst, 2019, 8(5): 380-394. DOI: 10.1016/j.cels.2019.04.003.

10.Hirst D, Térézol M, Cantini L, et al. MOTL: enhancing multi-omics matrix factorization with transfer learning[EB/OL]. (2024-03) [2025-04-10].https://doi.org/10.1101/2024. 03.22.586210.

11.Cai C, Wang S, Xu Y, et al. Transfer learning for drug discovery[J]. J Med Chem, 2020, 63(16): 8683-8694. DOI: 10.1021/acs.jmedchem.9b02147.

12.Ye Z, Yang Y, Li X, et al. An integrated transfer learning and multitask learning approach for pharmacokinetic parameter prediction[J]. Mol Pharm, 2019, 16(2): 533-541. DOI: 10.1021/acs.molpharmaceut.8b00816.

13.Turki T, Wei Z, Wang JTL. Transfer learning approaches to improve drug sensitivity prediction in multiple myeloma patients[J]. IEEE Access, 2017, 5: 7381-7393. DOI: 10.1109/ACCESS.2017.2696523.

14.Gore S, Meche B, Shao D, et al. DiseaseNet: a transfer learning approach to noncommunicable disease classification[J]. BMC Bioinformatics, 2024, 25(1): 107. DOI: 10.1186/s12859-024-05734-5.

15.Wang N, Zhang Y, Wang W, et al. How can machine learning and multiscale modeling benefit ocular drug development?[J]. Adv Drug Deliv Rev, 2023, 196: 114772. DOI: 10.1016/j.addr.2023.114772.

16.Guo W, Dong Y, Hao GF. Transfer learning empowers accurate pharmacokinetics prediction of small samples[J]. Drug Discov Today, 2024，29(4): 103946. DOI: 10.1016/j.drudis.2024.103946.

17.Haendel MA, Chute CG, Robinson PN. Classification, ontology, and precision medicine[J]. N Engl J Med, 2018, 379(15): 1452-1462. DOI: 10.1056/NEJMra1615014.

18.Boyd N, Dancey JE, Gilks CB, et al. Rare cancers: a sea of opportunity[J]. Lancet Oncol, 2016, 17(2): e52-e61. DOI: 10.1016/S1470-2045(15)00386-1.

19.Wang J, Ma F. Federated learning for rare disease detection: a survey[J/OL]. Rare Dis Orphan Drugs J, 2023, 2: 22. DOI: 10.20517/rdodj.2023.16.

20.Pati S, Baid U, Edwards B, et al. Federated learning enables big data for rare cancer boundary detection[J]. Nat Commun, 2022, 13(1): 7346. DOI: 10.1038/s41467-022-33407-5.

21.Chen B, Chen T, Zeng X, et al. DFML: dynamic federated meta-learning for rare disease prediction[J]. IEEE/ACM Trans Comput Biol Bioinform, 2024, 21(4): 880-889. DOI: 10.1109/TCBB.2023.3239848.

22.Li S, Cai T, Duan R. Targeting underrepresented populations in precision medicine: a federated transfer learning approach[J]. Ann Appl Stat, 2023, 17(4): 2970-2992. DOI: 10.1214/23-AOAS1747.

23.Alsentzer E, Li MM, Kobren SN, et al. Few shot learning for phenotype-driven diagnosis of patients with rare genetic diseases[J]. NPJ Digit Med, 2025, 8(1): 380. DOI: 10.1038/s41746-025-01749-1.

24.Dong D, Chung RYN, Chan RH, et al. Why is misdiagnosis more likely among some people with rare diseases than others? Insights from a population-based cross-sectional study in China[J]. Orphanet J Rare Dis, 2020, 15(1): 307. DOI: 10.1186/s13023-020-01587-2.

25.Wojtara M, Rana E, Rahman T, et al. Artificial intelligence in rare disease diagnosis and treatment[J]. Clin Transl Sci, 2023, 16(11): 2106-2111. DOI: 10.1111/cts.13619.

26.Lee J, Liu C, Kim J, et al. Deep learning for rare disease: a scoping review[J]. J Biomed Inform, 2022, 135: 104227. DOI: 10.1016/j.jbi.2022.104227.