http://biomed.nscc-gz.cn/sail/ Sun, 05 Apr 2026 06:59:46 +0000 FeedCreator 1.8 http://biomed.nscc-gz.cn/sail/_media/wiki:favicon.ico http://biomed.nscc-gz.cn/sail/ http://biomed.nscc-gz.cn/sail/labonly2:dokuwiki_setting?rev=1651498785&do=diff dokuwiki setting tpl/bootstrap3/assets/bootstrap/cerulean/bootstrap.min.css：颜色和字体大小 plugins/gallery/screen.less: portrait黑框 anonymous@undisclosed.example.com (Anonymous) labonly2 Mon, 02 May 2022 21:39:45 +0000 http://biomed.nscc-gz.cn/sail/labonly2:explainable_multiview_framework_for_dissecting_spatial_relationships_from_highly_multiplexed_data?rev=1668121837&do=diff scRNA-seq/Pharmogenomics * Integrating and formatting biomedical data as pre-calculated knowledge graph embeddings in the Bioteque * Gene expression based inference of cancer drug sensitivity * Connecting omics signatures and revealing biological mechanisms with iLINCS * Computational pharmacogenomic screen identifies drugs that potentiate the anti-breast cancer activity of statins * SpotClean adjusts for spot swapping in spatial transcriptomics data * ORIGINS: a protein net… anonymous@undisclosed.example.com (Anonymous) labonly2 Fri, 11 Nov 2022 07:10:37 +0000 http://biomed.nscc-gz.cn/sail/labonly2:%E7%BB%84%E4%BC%9A%E8%AE%BE%E7%BD%AE?rev=1718241051&do=diff 星生（星光生物）* 单细胞大模型预发表上线 联合华为、新格元两家各领域龙头公司共同研发的单细胞大模型CellFM正式预发表（biorxiv 2022.06.05.494905v2），该模型基于超1亿人类细胞训练（是其它模型两倍以上），通过华为升腾芯片训练得到超8亿参数的模型，是该类模型参数的8倍以上。该模型是此前与鹏城联合研发的72种脊椎动物RNA大模型SpliceBERT（约2千万参数，BIB 2024:bbae163）、和药物多模态预训练模型MIGA（6千万参数, Advanced Science 2024）之后的又一个预训练大模型。 未来，将充分利用超算算力和广泛的用户合作基础，在更多生物子领域开展预训练模型研究，从而最终构建面向生物的多尺度星生大模型，促进生物医药研究发展。… anonymous@undisclosed.example.com (Anonymous) labonly2 Thu, 13 Jun 2024 09:10:51 +0000