2026-10-06T23:20:06Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34073722 2026-10-05T17:52:04Z category_12 category_13 category_14 category_46 category_734 category_931 category_61 category_64 category_128 portal_5 item_type_3 month_year_10_2026

<p>Summary of the datasets utilized here.</p> Ritwik Ganguly (25317277) Sana Aafrine (25317280) Sk Md Mosaddek Hossain (4709248) Sumanta Ray (3505517) Cell Biology Genetics Molecular Biology Immunology Biological Sciences not elsewhere classified Information Systems not elsewhere classified Developmental Biology Cancer Hematology reduces mode dropping fidelity synthetic cells enhance downstream analyses raph -< b div >< p neighbour cell graph cell rna sequencing graph attention network coupling graph attention attention cell embeddings ra </ b ge </ b g </ b cell aware single across real scrna rare cell types real cells cell rna downstream single aware single </ b k </ respecting cell cell structure cell data cell clustering garage </ type proportions samples generated prioritize nodes prior noise often compounded likely represent generator ’ existing simulators data manifold corresponding software clustering compared class imbalance central challenge attentive gan art baselines adversarial generation >- nearest >&# 8216 <div><p>A central challenge in downstream single-cell RNA sequencing (scRNA-seq) analysis is the high-dimensional, small-sample (HDSS) regime, often compounded by class imbalance from rare cell types. These factors hinder robust feature (gene) selection and cell clustering and limit the realism of samples generated by existing simulators. We introduce <i>GARAGE</i>, a <b>G</b>raph-<b>A</b>ttentive <b>RA</b>re-cell aware single-cell data <b>GE</b>neration that augments the generator’s input with a small, attention-weighted <i>‘leakage’</i> of real cells in addition to prior noise. Specifically, we build a <i>k</i>-nearest-neighbour cell graph and use a graph attention network (GAT) to prioritize nodes that likely represent under-sampled (rare) subpopulations; these high-attention cell embeddings are injected into the generator input to steer synthesis toward biologically plausible regions of the data manifold while respecting cell-type proportions. This attention-guided leakage accelerates training, reduces mode dropping, and yields realistic synthetic cells that preserve rare-cell structure. Across real scRNA-seq benchmarks, GARAGE improves downstream feature selection and clustering compared with state-of-the-art baselines. In summary, GARAGE directly addresses HDSS and rarity in scRNA-seq by coupling graph attention with adversarial generation to produce high-fidelity synthetic cells that enhance downstream analyses. The corresponding software is available at: <a href="https://github.com/RitwikGanguly/GARAGE" target="_blank">https://github.com/RitwikGanguly/GARAGE</a>.</p></div> 2026-10-05T17:51:54Z Dataset Dataset 10.1371/journal.pcbi.1014601.t001 https://figshare.com/articles/dataset/_p_Summary_of_the_datasets_utilized_here_p_/34073722 CC BY 4.0