i4 -- raw regression output with regards to flight

Microbiome associations with spaceflight across classifiers, sequencing modalities, and microbiome features. Raw outputs are stored as .rds files. CSV files are the "categorized" (i.e., parsed) regression based on each feature's trajectory relative to flight (transiently increased, transiently decreased, persistently increased, persistently decreased, etc). Brief methods are as follows:<br><br>We grouped microbial features associated with flight into six different categories. These were determined due to the fact that our model contained a categorical variable encoding a sample’s timing relative to flight: whether it was taken before, during, or afterwards. Since the modeling reference group was “MID-FLIGHT,” meaning that the interpretation of any coefficients would be directionally oriented relative to mid-flight microbial feature abundances. As a result, we were able to categorize features based on the jointly considered direction of association and significance for the “PRE-FLIGHT” and “POST-FLIGHT” levels of this variable. The below listed categories are all included in the association summaries provided in Supplementary Table 3.Transient increase in-flight – negative coefficient on the PRE-FLIGHT variable level, negative coefficient on the POST-FLIGHT variable, statistically significant for both Transient increase in-flight (low priority) – negative coefficient on the PRE-FLIGHT variable level, negative coefficient on the POST-FLIGHT variable, statistically significant for at least one of the twoTransient decrease in-flight – positive coefficient on the PRE-FLIGHT variable level, positive coefficient on the POST-FLIGHT variable level, statistically significant for both Transient decrease in-flight (low priority) – positive coefficient on the PRE-FLIGHT variable level, positive coefficient on the POST-FLIGHT variable level, statistically significant for at least one of the twoPotential persistent increase – negative coefficient on the PRE-FLIGHT variable level, positive coefficient on the POST-FLIGHT variable level, statistically significant for at least one of the twoPotential persistent decrease – positive coefficient on the PRE-FLIGHT variable level, negative coefficient on the POST-FLIGHT variable level, statistically significant for at least one of the two<br><br>

本数据集聚焦不同分类器、测序模式及微生物组特征下，微生物组与航天飞行的关联关系。原始输出结果以.rds文件格式存储。CSV文件为经分类（即解析后）的回归分析结果，基于各特征相对于航天飞行的轨迹进行划分，涵盖短暂升高、短暂降低、持续升高、持续降低等类别。研究方法简述如下：

我们将与航天飞行相关的微生物组特征划分为六大类别。分类依据来自模型中设置的分类变量，该变量编码样本采集相对于航天飞行的时间节点：即飞行前（PRE-FLIGHT）、飞行中（MID-FLIGHT）与飞行后（POST-FLIGHT）。由于建模参考组为"飞行中（MID-FLIGHT）"，因此所有系数的解读均以飞行中微生物组特征的丰度为参照方向。据此，我们可结合该变量"飞行前（PRE-FLIGHT）"与"飞行后（POST-FLIGHT）"两个水平的关联方向与统计学显著性，对微生物特征进行分类。以下列出的类别均包含于补充表3的关联总结中：
1. 飞行中短暂升高：飞行前变量水平系数为负、飞行后变量水平系数为负，且二者均具有统计学显著性
2. 飞行中短暂升高（低优先级）：飞行前变量水平系数为负、飞行后变量水平系数为负，且至少其中一个具有统计学显著性
3. 飞行中短暂降低：飞行前变量水平系数为正、飞行后变量水平系数为正，且二者均具有统计学显著性
4. 飞行中短暂降低（低优先级）：飞行前变量水平系数为正、飞行后变量水平系数为正，且至少其中一个具有统计学显著性
5. 潜在持续升高：飞行前变量水平系数为负、飞行后变量水平系数为正，且至少其中一个具有统计学显著性
6. 潜在持续降低：飞行前变量水平系数为正、飞行后变量水平系数为负，且至少其中一个具有统计学显著性