跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2610.33307 · v3

Tree Depth and Sample-Quality-Associated Features: Two Hidden Factors in Boruta Selection for RNA-Seq Data Classification

Tareque Mohmud ChowdhuryAbu Raihan Mostofa Kamal
cs.LG
首次提交 2026-10-24 05:19
版本更新 2026-10-29 15:56
雷达观测 2026-10-30 13:28
arXiv ↗PDF ↗

摘要

Random Forest--based selectors such as Boruta depend on tree depth, yet Boruta's most used implementation, BorutaPy, recommends depths of 3 to 7 without published evidence. We tested the effect of depth using pan-cancer data from The Cancer Genome Atlas (10,469 tumors, 33 cancer types) and breast cancer PAM50 subtyping (981 tumors, 5 subtypes). Label-dependent steps were fitted inside training folds, effects were measured against seed-to-seed variation, and each task had a locked test set. In pan-cancer data, depth 5 changed the selection far more than a new seed did: the top-500 Jaccard overlap with unlimited depth was 0.37, against 0.84 (depth 5) and 0.93 (unlimited) between seeds. Depth-5 trees confirmed 472 instead of 1,659 genes, covered fewer large-class markers (16% vs 62%) and gave less accurate 50-gene panels (balanced accuracy 0.809 vs 0.889). The effect vanished once the cap reached 61 to 76% of the median depth of unrestricted trees, a depth set mainly by the number of classes, not sample size. In PAM50 subtyping (median depth 12) the effect was small, and leaving depth uncapped cost no accuracy in either task. The effect replicated in a second cross-validation repeat and under permutation importance. About a third of pre-filtered candidates correlated within class with the number of detected genes, a sample-quality measure; a within-class filter removed them without loss of test accuracy (0.943 vs 0.942 balanced accuracy, 1,000 genes). We recommend uncapped trees, or a cap above 75% of the realized median depth, and a within-class quality filter before Boruta.

作者备注：19 Pages, 6 Figures, 4 Tables

查看 arXiv 原文 ↗阅读 PDF ↗

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存，登录后共享已有结果。

版本差异

仅对比实际观测的元数据，不代表 PDF 正文或实验变化。版本可能不连续；带删除线为移除，下划线为新增。

摘要变化

Boruta is a widely usedRandom methodForest--based forselectors selectingsuch geneas panelsBoruta fromdepend RNAon sequencingtree datadepth, andyet itsBoruta's most used implementation recommends, RandomBorutaPy, Forestrecommends treesdepths of depth 3 to 7, anwithout untestedpublished settingevidence. We tested whether this setting changes whichthe geneseffect areof selected,depth using pan-cancer data from The Cancer Genome Atlas (10,469 tumors, 33 cancer types) and breast cancer PAM50 subtyping (981 tumors, 5 subtypes). A leakageLabel-free protocol fitteddependent everysteps label-dependentwere stepfitted inside training folds, measuredeffects everywere effectmeasured against seed-to-seed variation, and heldeach outtask had a locked test set per task. In pan-cancer data, shallowdepth trees5 changed the selection far more than a new seed did: the top-500 Jaccard overlap with unlimited-depth selectionsdepth was 0.37 at depth 5, against 0.84 to(depth 5) and 0.93 (unlimited) between seeds at depth 5 or unlimited depth. ShallowDepth-5 trees confirmed 3.5472 timesinstead fewerof 1,659 genes, missedcovered fewer large-class markers of(16% largevs classes62%) and producedgave less accurate 50-gene panels (balanced accuracy 0.809 vs 0.889). The effect vanished once the cap reached about 6061 to 7576% of the median depth thatof unrestricted trees, growa to,depth set mainly by the number of classes, not sample size. In PAM50 subtyping (median depth 12) the effect was small, and leaving depth uncapped cost no accuracy in either task. The effect replicated in a second cross-validation repeat and under permutation importance. About a third of pre-filtered candidates alsocorrelated trackedwithin sampleclass qualitywith withinthe classnumber of detected genes, a sample-quality measure; a within-class filter removed them without loss of test accuracy (balanced accuracy 0.943 vs 0.942 withbalanced accuracy, 1,000 genes). We recommend uncapped trees, or a cap chosenabove from75% of the realized treemedian depth, and a within-class quality filter before Boruta.

观测到的版本

相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。

v3 · 2026-10-29 15:56 · 最新观测版本

Tree Depth and Sample-Quality-Associated Features: Two Hidden Factors in Boruta Selection for RNA-Seq Data Classification

Random Forest--based selectors such as Boruta depend on tree depth, yet Boruta's most used implementation, BorutaPy, recommends depths of 3 to 7 without published evidence. We tested the effect of depth using pan-cancer data from The Cancer Genome Atlas (10,469 tumors, 33 cancer types) and breast cancer PAM50 subtyping (981 tumors, 5 subtypes). Label-dependent steps were fitted inside training folds, effects were measured against seed-to-seed variation, and each task had a locked test set. In pan-cancer data, depth 5 changed the selection far more than a new seed did: the top-500 Jaccard overlap with unlimited depth was 0.37, against 0.84 (depth 5) and 0.93 (unlimited) between seeds. Depth-5 trees confirmed 472 instead of 1,659 genes, covered fewer large-class markers (16% vs 62%) and gave less accurate 50-gene panels (balanced accuracy 0.809 vs 0.889). The effect vanished once the cap reached 61 to 76% of the median depth of unrestricted trees, a depth set mainly by the number of classes, not sample size. In PAM50 subtyping (median depth 12) the effect was small, and leaving depth uncapped cost no accuracy in either task. The effect replicated in a second cross-validation repeat and under permutation importance. About a third of pre-filtered candidates correlated within class with the number of detected genes, a sample-quality measure; a within-class filter removed them without loss of test accuracy (0.943 vs 0.942 balanced accuracy, 1,000 genes). We recommend uncapped trees, or a cap above 75% of the realized median depth, and a within-class quality filter before Boruta.

查看此版本解读 →查看此版本原文 ↗
v2 · 2026-10-27 15:44

Tree Depth and Sample-Quality-Associated Features: Two Hidden Factors in Boruta Selection for RNA-Seq Data Classification

Boruta is a widely used method for selecting gene panels from RNA sequencing data, and its most used implementation recommends Random Forest trees of depth 3 to 7, an untested setting. We tested whether this setting changes which genes are selected, using pan-cancer data from The Cancer Genome Atlas (10,469 tumors, 33 cancer types) and breast cancer PAM50 subtyping (981 tumors, 5 subtypes). A leakage-free protocol fitted every label-dependent step inside training folds, measured every effect against seed-to-seed variation, and held out a locked test set per task. In pan-cancer data, shallow trees changed the selection far more than a new seed did: the top-500 overlap with unlimited-depth selections was 0.37 at depth 5, against 0.84 to 0.93 between seeds at depth 5 or unlimited depth. Shallow trees confirmed 3.5 times fewer genes, missed markers of large classes and produced less accurate 50-gene panels (balanced accuracy 0.809 vs 0.889). The effect vanished once the cap reached about 60 to 75% of the depth that unrestricted trees grow to, set mainly by the number of classes, not sample size, and leaving depth uncapped cost no accuracy in either task. The effect replicated in a second cross-validation repeat and under permutation importance. About a third of pre-filtered candidates also tracked sample quality within class; a within-class filter removed them without loss of test accuracy (balanced accuracy 0.943 vs 0.942 with 1,000 genes). We recommend uncapped trees, or a cap chosen from the realized tree depth, and a within-class quality filter before Boruta.

查看此版本解读 →查看此版本原文 ↗