[2] viXra:2611.0047 [pdf] submitted on 2026-11-15 01:47:35
Authors: Eric Zhuang, Manoj Kandpal
Comments: 32 Pages.
Ribosome profiling sequences messenger RNA (mRNA) fragments protected by ribosomes. Each sequenced fragment is represented as a read. Where a read aligns shows the region occupied by a ribosome, but it does not directly show which nucleotide lies at the ribosome's peptidyl site (P site). Identifying that nucleotide is necessary for analyses of individual codons, the three-nucleotide units read during translation. Existing P-site predictors require different inputs, define positions differently, report either one position per read or a signal summarized across many reads, and vary in whether they use information from the experimental sample being evaluated. These differences make fair comparison difficult.We developed TabRiboICL, a novel predictor that assigns one P-site position to each read. It evaluates 13 possible positions located 8—20 nucleotides from the read's 5u2032 end. The published TabICL machine-learning model [10] scores each possible position. TabRiboICL combines each score with how often the corresponding distance occurred among similar training reads with expected P-site positions and then selects one P-site position. We also developed a benchmarking framework that converts the different forms of output to the same definition of a read-level P-site position and evaluates every predictor on the same reads.The benchmark includes 8 public datasets, 31 ribosome-profiling samples from Drosophila melanogaster and Saccharomyces cerevisiae, and 11 predictors. An expected P-site position was assigned to a read near the annotated beginning or end of a protein-coding region only when exactly one of the 13 possible positions matched the relevant boundary and fell on the first nucleotide of an annotated codon. Because these expected positions were inferred rather than directly observed, we call them proxy labels. Exact accuracy, the primary measure, required a prediction to match its proxy label. Accuracy within one nucleotide was the secondary measure.During the held-out dataset experiment, TabRiboICL had the highest observed mean exact accuracy after every dataset was given equal weight: 37.6% (95% bootstrap confidence interval, 32.5%—43.8%). The interval was calculated by repeatedly redrawing datasets, samples, and reads. TabRiboICL had the highest observed exact accuracy in 5 of 8 datasets; Scikit-ribo had the highest in two datasets and LightGBM in one. The framework also characterized predictor performance across datasets, organisms, and recorded protocol categories, providing a consistent, well-documented basis for comparing P-site predictors.
Category: Quantitative Biology
[1] viXra:2611.0028 [pdf] submitted on 2026-11-10 00:51:08
Authors: Tianjiao Wang
Comments: 12 Pages.
Prognostic transcriptomic signatures are commonly evaluated in bulk tumour tissue, where measured expression reflects malignant cells together with stromal and immune compartments. Tumour purity may therefore be associated with both a signature score and survival, complicating the interpretation of an unadjusted prognostic coefficient. We conducted a pan-cancer audit of published prognostic transcriptomic signatures across 12 The Cancer Genome Atlas (TCGA) cohorts, examining how purity adjustment changes their associations with overall survival. The inventory comprised54 eligible published signatures, six reference gene sets, and two additional published signatures excluded for insufficient gene coverage. One included ovarian lncRNA signature could not be scored in its intended cohort because a component gene had zero variance,yielding 53 published signatures and six reference sets (59 distinct scored sets) across 129 signature—cohort pairs. The primary analysisincluded 5,417 patients and 1,939 deaths. After false-discovery-rate correction, 97 of 129 pairs showed an association between signature score and ABSOLUTE purity. The median absolute coefficient change after purity adjustment was 0.022; four pairs were classified as attenuated, 13 as amplified, and 112 as not distinguishablefrom zero. All four attenuated pairs were reference sets rather than published prognostic signatures. The magnitude and classificationof coefficient changes varied across purity estimators, highlighting the importance of estimator choice when evaluating prognostictranscriptomic signatures. These findings describe conditional associations and do not imply a causal effect of tumour purity.
Category: Quantitative Biology