Dataset

mRNABench:mrl-sample-egfp-m1pseudo

mRNABench:mrl-sample-egfp-m1pseudo

Not in the Core Set 10 of the 93 cleared all five gates

Task: Regression, Translation Efficiency Prediction
Level: RNA
Species: Homo sapiens
Sequences: 318,468 across 3 splits
Sequence length: 855 to 855
Licence: No restrictions (GEO deposition)

Read the 5 full reports for this dataset Every distribution, probe fit and overlapping sequence behind the numbers below

Fail worst flag across the ten checks on train, this dataset's reference split
0 of the ten checks returned Fail on that split 1 returned Warning, 0 could not be scored
0.667 AUROC from the strongest single sequence feature, which is Per sequence dinucleotide content A lower bound on what a model could recover from the sequences alone
100.00% of one split's sequences were also present in another, at worst The largest of 2 split comparisons: train vs val
0.857≥ a lower bound AUROC when all those features are read together rather than one at a time Measured on a held-out split. This is what Core Set membership turns on

Where this sits in the corpus

Stronger than 47 of the 93 probed datasets
Higher than 57 of the 92 jointly scored datasets
More overlap than 87 of the 92 scanned datasets

A rank among the datasets this run measured, not among benchmarks in general.

What the checks returned

Worst flag across the ten checks on the train split: Fail

Of the nine checks that measure a single split, 9 ran on train and 9 were scored: 0 returned Fail, 1 returned Warning, 0 returned Unknown. The leakage scan is counted in the flag above but not in these figures, because it compares pairs of splits rather than measuring one.

Every rollup figure on this page was measured on train, which is this dataset's reference split — not a maximum over its splits.

The strongest single feature is Per sequence dinucleotide content, which reaches AUROC 0.667 on the train split from that feature alone. That is a lower bound: a stronger feature, or a model, may recover more.

The leakage scan found that 100.00% of the sequences in val were also present in train; the same comparison the other way puts 99.90% of train in val. That is the largest of the two split comparisons the scan made here.

What each of the ten checks measures · What a flag is, and what it does not say · Why one split is quoted

The reports behind those flags

The full report genomic-benchmarks-qc rendered for each split and each split pair, with every distribution, every probe fit, and every individually-listed overlapping sequence behind the numbers on this page. Each opens inside this site, and the file the tool wrote is one click further on.

train · sequence class distributions
val · sequence class distributions
test · sequence class distributions
train vs test leakage scan
train vs val leakage scan

Worked examples of a report

The same features, read together

Every flag above is one feature on its own. This is all of them at once: length, GC, mononucleotide and dinucleotide content and the two per-position profiles, handed to a single model and scored on a held-out split. Nothing is added that the checks do not already look at — the only change is that they are read jointly.

0.857 AUROC from all the checks’ features at once, measured on the test split 95% interval 0.854 to 0.860, over 47,769 held-out sequences

The strongest single feature reaches 0.667 on its own.

This figure is a lower bound rather than a point value: training pool subsampled. It can only be too low, so what these features recover together may be more than the number says.

How it was measured, why it exists, and the six datasets it changes the reading of

The nine per-split checks

One row per split, one column per check. The leakage scan is not here because it compares split pairs rather than measuring one split; it has its own section below.

The nine class-distribution checks on each of mRNABench:mrl-sample-egfp-m1pseudo’s 3 splits. The rollup reports the train row.

Table view

Flag and, where measured, the strongest probe AUROC per check per split.
Split	GC content	Nucleotide content	Dinucleotide content	Sequence lengths	Per position nucleotide	Per position from 3′	Duplicates within label	Duplicates across labels	Unique bases
train	Pass 0.521	Pass 0.582	Warning 0.667	Pass 0.500	Pass 0.541	Pass 0.541	Pass	Pass	Pass
val	Pass 0.523	Pass 0.585	Warning 0.666	Pass 0.500	Pass 0.541	Pass 0.541	Pass	Pass	Pass
test	Pass 0.525	Pass 0.587	Warning 0.668	Pass 0.500	Pass 0.540	Pass 0.540	Pass	Pass	Pass

Each split, the flag it rolled up to, its strongest probe, and the report every flag on that row was read out of.
Split	Sequence columns	Worst flag	Checks scored	Strongest probe	AUROC	Report
`train` reference	`sequence`	Warning	9	Per sequence dinucleotide content	0.667	sequence
`val`	`sequence`	Warning	9	Per sequence dinucleotide content	0.666	sequence
`test`	`sequence`	Warning	9	Per sequence dinucleotide content	0.668	sequence

How far each probe gets

The AUROC each of the six probe checks reaches on each split, from that one feature alone. The axis starts at 0.50 because that is what a feature carrying no class information returns.

What each single-feature probe recovered, on each split. Where a dataset has several sequence columns the strongest is shown; the table view lists every value.

The axis starts at 0.50 because that is what a probe recovering nothing would read. A value is a lower bound: it is what one feature reached, not the most that could be reached.

Table view

Probe AUROC per check per split.
Check	train	val	test
Per sequence GC content	0.521	0.523	0.525
Per sequence nucleotide content	0.582	0.585	0.587
Per sequence dinucleotide content	0.667	0.666	0.668
Sequence lengths	0.500	0.500	0.500
Per position nucleotide content	0.541	0.541	0.540
Per position reversed nucleotide content	0.541	0.541	0.540

Split comparisons

For every pair of splits the scan compared, the share of each split whose sequences also appear in the other. Each bar is coloured by the split it counts.

The two directions of a pair are different statements. A held-out split largely contained in train is the one that bears on whether a score on it is a held-out score.

Table view

Exact-sequence overlap, both directions, per split pair.
Comparison	Share	Flag
train also in val	99.90%	Fail
val also in train	100.00%	Fail
train also in test	99.90%	Fail
test also in train	100.00%	Fail

Each pair, in both directions. The two percentages differ whenever the two splits differ in size, so both are shown rather than the larger one.
Comparison	Flag	Of the first split	Of the second
`val` in `train`	Fail	100.00%	99.90%
`test` in `train`	Fail	100.00%	99.90%

Licence, credit, and the data itself

What this data may be used for is its source's decision and not this project's. It is set down here as the curation read it, with the evidence, so that a reader can check the reading rather than take it.

Licence: No restrictions (GEO deposition)
Credit: 10.1101/2025.07.05.662870 — the collection's

Each split's file, and where it is served from:

train — mrl-sample-egfp-m1pseudo_train.csv.gzhuggingface.co
val — mrl-sample-egfp-m1pseudo_val.csv.gzhuggingface.co
test — mrl-sample-egfp-m1pseudo_test.csv.gzhuggingface.co

This licence permits redistribution, so the split files may be republished under its terms rather than only fetched from the links above. mRNABench collected this data from GEO, series GSE114002 (Sample et al. 2019), which is public. NCBI's data usage policy states: "NCBI itself places no restrictions on the use or distribution of the data contained therein. Nor do we accept data when the submitter has requested restrictions on reuse or redistribution." Deposition is therefore evidence that no redistribution restriction was requested. Recorded with the qualification that this is an inference from the repository's intake policy rather than a grant from the authors: the same policy notes submitters may still hold copyright, and NCBI cannot give permission on their behalf. The paper states no data availability terms of its own.

How to cite the run these numbers come from

Nearby

Nearest by name in mRNABench: mRNABench:mrl-sample-designed, mRNABench:mrl-sample-egfp-pseudo, mRNABench:mrl-sample-egfp-unmod, mRNABench:mrl-sample-mcherry, mRNABench:mrl-sample-varying, mRNABench:mrl-sugimoto, mRNABench:rnahl-human, mRNABench:rnahl-mouse · and 2 more