MDSINE2 Cross-Validation run (Dysbiotic cohort)
收藏资源简介：
MDSINE2 Inference Analysis files. This archive contains all output files from the cross-validation inference (with comparator analysis included) for Dysbiotic cohort. (Both healthy and dysbiotic cohorts are required to run the Jupyter Notebook (`fig4_semisynthetic_v2_cache.ipynb`) on our MDSINE2_Paper repo. For the Healthy cohort files, refer to https://zenodo.org/records/16905647. For the full project/source pipeline, refer to https://github.com/gerberlab/MDSINE2_Paper. Please note that, unlike the healthy cohort cross-validation record (zenodo.16905647) this file does not include MDSINE2's numpy-array raw outputs (HDF5-formatted MCMC trace data) due to file-size limitations on Zenodo. We have, however, included the pre-processed forward-simulation files ready to be used with the cross-validation evaluation scripts (`scripts/cross_validation/evaluate_errors.sh`). If you need the raw MCMC outputs, you can (1) get in touch with the authors directly (tegibson@bwh.harvard.edu, ykim78@bwh.harvard.edu) or (2) re-run the cross-validation analysis from scratch using the provided scripts in the github repository. The archive here was created using the command, which excludes HDF5 files: tar --zstd --exclude='*.hdf5' -cvf cross_validation_uc.tar.zst datasets/gibson/uc/cross_validation To unpack this file, please use the following command: tar -I zstd -xvf cross_validation_uc.tar.zst These files should be unpacked and placed in the paper repository directory (wherever you did `git clone`), so that the `datasets` directory is directly inside `MDSINE2_Paper` repository directory. ------------ Related zenodo records: https://doi.org/10.5281/zenodo.8208502 -- Full MDSINE2 inference on Healthy cohort https://doi.org/10.5281/zenodo.16905647 -- MDSINE2 Cross-Validation run (Healthy cohort) https://doi.org/10.5281/zenodo.16915311 -- MDSINE2 semisynthetic dataset
MDSINE2 推断分析文件。本归档包包含失调菌群队列（Dysbiotic cohort）的交叉验证推断（含对照分析）的全部输出文件。运行本团队MDSINE2_Paper仓库中的Jupyter Notebook（`fig4_semisynthetic_v2_cache.ipynb`）时，需同时使用健康菌群队列与失调菌群队列的相关文件；健康菌群队列的文件可访问https://zenodo.org/records/16905647获取。 如需获取完整的项目与源代码流程，请访问https://github.com/gerberlab/MDSINE2_Paper。 请注意，与健康菌群队列的交叉验证记录（zenodo.16905647）不同，受限于Zenodo的文件大小限制，本归档包未包含MDSINE2的numpy数组格式原始输出（即HDF5格式的马尔可夫链蒙特卡洛（Markov Chain Monte Carlo, MCMC）追踪数据）。不过我们已将可直接用于交叉验证评估脚本（`scripts/cross_validation/evaluate_errors.sh`）的预处理正向模拟文件包含在内。若您需要原始MCMC输出结果，可通过以下两种方式获取：（1）直接联系本文作者（邮箱：tegibson@bwh.harvard.edu、ykim78@bwh.harvard.edu）；（2）使用GitHub仓库中提供的脚本重新从头运行交叉验证分析。 本归档包通过以下命令创建，该命令已排除HDF5格式文件： `tar --zstd --exclude='*.hdf5' -cvf cross_validation_uc.tar.zst datasets/gibson/uc/cross_validation` 如需解压该文件，请使用以下命令： `tar -I zstd -xvf cross_validation_uc.tar.zst` 请将解压后的文件放置于论文仓库目录（即您执行`git clone`的目录）中，确保`datasets`目录直接位于`MDSINE2_Paper`仓库目录下。 --- 相关Zenodo记录： https://doi.org/10.5281/zenodo.8208502 —— 健康菌群队列的完整MDSINE2推断结果 https://doi.org/10.5281/zenodo.16905647 —— 健康菌群队列的MDSINE2交叉验证运行结果 https://doi.org/10.5281/zenodo.16915311 —— MDSINE2半合成数据集
