Uncertainty-aware benchmarking reveals ambiguous transcripts in mRNA–lncRNA classification
Abstract
Background Long non-coding RNAs (lncRNAs) have gained significant attention in recent years, yet distinguishing them from protein-coding transcripts remains challenging. Indeed, many lncRNAs share mRNA-like processing and existing sequence-derived signals do not fully capture the coding/non-coding boundary. Recent GENCODE annotation efforts revealed tens of thousands of novel lncRNA sequences as well as the reclassification of some lncRNAs into the protein-coding class, highlighting the need to better characterize transcript features associated with classification uncertainty and errors. Results We performed uncertainty-aware benchmarking by retraining and evaluating eight transcript classifiers under a controlled protocol on a label-stable GENCODE v46–v47 subset. Beyond conventional model evaluation metrics, we quantified inter-tool agreement and entropy-based uncertainty to stratify transcripts into low and high uncertainty groups. Although aggregate performance was high, ~ 45% of transcripts showed inter-tool discordance, particularly among lncRNAs. To expand standard sequence and ORF-derived signals, we incorporated repeat-derived features and non-canonical secondary structure from both mature transcripts and gene bodies. Feature analyses linked low-uncertainty predictions to strong coding-like signals, whereas high-uncertainty profiles exhibited mixed signatures. Alongside classical predictors in global importance analyses, repeat-derived features appeared as main contributors. Conclusions By combining controlled benchmarking with transcript-level agreement and uncertainty stratification, together with extended feature profiling, we identified patterns associated with classifier disagreement and misclassification. This novel framework provides practical guidance for interpreting predictions, motivating the development of more robust coding/non-coding classifiers, while also shedding light on the sequence properties that distinguish lncRNA sequences.
Related
- Long-read RNA sequencing reveals full-length HERV transcripts and novel ERVE-4 isoforms with potential immunotherapeutic relevance in clear cell renal cell carcinoma
- Cross-task benchmarking reveals variable transferability of deep learning models across molecular quantitative trait loci.
- Benchmarking single-cell format conversion tools reveals interoperability gaps and introduces CrossCell for reliable cross-ecosystem data exchange.
- Interlaboratory benchmarking of differential scanning fluorimetry reveals sources of variability in protein thermal stability parameters
- VRK1 regulates SRRM2 phosphorylation to promote the splicing of transcripts encoding DNA damage repair factors
BibTeX
@misc{garciaruano2026uncertaintyaware,
title = {{Uncertainty-aware benchmarking reveals ambiguous transcripts in mRNA–lncRNA classification}},
author = {Daniel Garcia-Ruano and Mikaël Georges and Saswat K. Mohanty and Rahma Baaziz and Kateryna D. Makova and Macha Nikolski and Domitille Chalopin},
howpublished = {Research Square},
year = {2026},
month = oct,
doi = {10.21203/rs.3.rs-11362934/v1},
url = {https://doi.org/10.21203/rs.3.rs-11362934/v1},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Uncertainty-aware benchmarking reveals ambiguous transcripts in mRNA–lncRNA classification
AU - Daniel Garcia-Ruano
AU - Mikaël Georges
AU - Saswat K. Mohanty
AU - Rahma Baaziz
AU - Kateryna D. Makova
AU - Macha Nikolski
AU - Domitille Chalopin
PY - 2026
DA - 2026/10/29
PB - Research Square
DO - 10.21203/rs.3.rs-11362934/v1
UR - https://doi.org/10.21203/rs.3.rs-11362934/v1
AB - Background Long non-coding RNAs (lncRNAs) have gained significant attention in recent years, yet distinguishing them from protein-coding transcripts remains challenging. Indeed, many lncRNAs share mRNA-like processing and existing sequence-derived signals do not fully capture the coding/non-coding boundary. Recent GENCODE annotation efforts revealed tens of thousands of novel lncRNA sequences as well as the reclassification of some lncRNAs into the protein-coding class, highlighting the need to better characterize transcript features associated with classification uncertainty and errors. Results We performed uncertainty-aware benchmarking by retraining and evaluating eight transcript classifiers under a controlled protocol on a label-stable GENCODE v46–v47 subset. Beyond conventional model evaluation metrics, we quantified inter-tool agreement and entropy-based uncertainty to stratify transcripts into low and high uncertainty groups. Although aggregate performance was high, ~ 45% of transcripts showed inter-tool discordance, particularly among lncRNAs. To expand standard sequence and ORF-derived signals, we incorporated repeat-derived features and non-canonical secondary structure from both mature transcripts and gene bodies. Feature analyses linked low-uncertainty predictions to strong coding-like signals, whereas high-uncertainty profiles exhibited mixed signatures. Alongside classical predictors in global importance analyses, repeat-derived features appeared as main contributors. Conclusions By combining controlled benchmarking with transcript-level agreement and uncertainty stratification, together with extended feature profiling, we identified patterns associated with classifier disagreement and misclassification. This novel framework provides practical guidance for interpreting predictions, motivating the development of more robust coding/non-coding classifiers, while also shedding light on the sequence properties that distinguish lncRNA sequences.
SN - 2693-5015
ER -