Summary

Basic Descriptive Statistics
! Data Leakage
Documentation GitHub
Find the shortcut. Learn the biology.
Automated quality control for genomic machine learning datasets: scores the biases, duplicates and data leakage a classifier could exploit before you train on it.
Searching test set H3K27me3_test.csv.gz against train set H3K27me3_train.csv.gz
Data Leakage:!WarningHow a flag is decided →
gb-qc v1.0.0 · run 2026-09-15 22:31:24

Basic Descriptive Statistics

Filename H3K27me3_test.csv.gz H3K27me3_train.csv.gz
Number of sequences 3000 30000
Minimum length 1000 1000
Mean length 1000.00 1000.00
Maximum length 1000 1000
!

Data Leakage

Every test sequence searched against the train set with MMseqs2, and scored on its best hit. Similarity is min(query coverage, target coverage) × percent identity: how much of the shorter sequence the alignment spans, times how much of that matches. The histogram is the distribution of those best hits.

The flag is the percentage of test sequences at or above the threshold: Pass at 0%, Warning below 2%, Fail at 2% or more. A test sequence that near-duplicates a training one is one the model has already seen, so the accuracy it earns there is not evidence. What to do about it →

The panel under the histogram lists those alignments, up to the first 100; every one of them, listed or not, is exported to mmseqs/mmseqs2_search_result.tsv beside this report. Its columns are:

Query (Q)
Identifier of the test sequence: seq_<i>_test, where i is its 0-based position in the input test file.
Target (T)
Identifier of the train sequence: seq_<i>_train, where i is its 0-based position in the input train file.
Q Cov.
Fraction of the query sequence covered by the alignment (0–1).
T Cov.
Fraction of the target sequence covered by the alignment (0–1).
% Identity
Percent identical aligned positions in the aligned region.
% Similarity
Similarity score used for leakage detection, calculated as min(Q Cov., T Cov.) × % Identity.
E-value
MMseqs2 E-value (lower is more "significant").
Alignment
Click “Show” to expand the alignment visualisation for a row.
Filename H3K27me3_test.csv.gz H3K27me3_train.csv.gz
Data Leakage (percentage) 0.93% 0.10%
Data Leakage (count) 28 31
Similarity Histogram
47 high-similarity alignments
Query (Q) Target (T) Q Cov. T Cov. % Identity % Similarity E-value Alignment
seq_1306_test seq_28869_train 0.99 0.99 97.6 96.82 0.00e+00
seq_1584_test seq_3804_train 0.99 0.99 97.4 96.43 0.00e+00
seq_880_test seq_8633_train 0.99 0.99 97.2 96.03 0.00e+00
seq_2375_test seq_19661_train 0.98 0.98 97.3 95.55 0.00e+00
seq_730_test seq_22435_train 0.99 0.99 96.4 95.24 0.00e+00
seq_1584_test seq_9306_train 0.97 0.97 98.3 95.15 0.00e+00
seq_356_test seq_6590_train 0.98 0.98 96.7 94.86 0.00e+00
seq_1627_test seq_5209_train 0.99 0.99 95.9 94.85 0.00e+00
seq_356_test seq_19661_train 0.97 0.97 97.5 94.48 0.00e+00
seq_2501_test seq_7309_train 0.98 0.98 96.4 94.09 0.00e+00
seq_2375_test seq_6590_train 0.97 0.97 97.0 93.90 0.00e+00
seq_2375_test seq_29301_train 0.97 0.97 96.9 93.61 0.00e+00
seq_990_test seq_18055_train 0.98 0.98 95.6 93.50 0.00e+00
seq_2353_test seq_7309_train 0.96 0.96 97.1 93.31 0.00e+00
seq_534_test seq_7309_train 0.96 0.96 96.7 93.22 0.00e+00
seq_1590_test seq_21732_train 0.97 0.98 95.8 93.21 0.00e+00
seq_880_test seq_24763_train 0.97 0.97 96.1 93.02 0.00e+00
seq_1927_test seq_26580_train 0.98 0.98 94.5 92.89 0.00e+00
seq_2595_test seq_5209_train 0.96 0.96 96.3 92.64 0.00e+00
seq_304_test seq_4326_train 0.95 0.95 97.0 92.44 0.00e+00
seq_356_test seq_10235_train 0.96 0.96 96.2 92.35 0.00e+00
seq_297_test seq_5209_train 0.95 0.96 96.5 92.16 0.00e+00
seq_356_test seq_29301_train 0.95 0.95 96.7 92.16 0.00e+00
seq_2104_test seq_28396_train 0.97 0.96 95.4 92.06 0.00e+00
seq_2375_test seq_10868_train 0.94 0.94 97.4 91.85 0.00e+00
seq_131_test seq_26345_train 0.98 0.99 93.3 91.81 0.00e+00
seq_730_test seq_18055_train 0.96 0.96 95.8 91.68 0.00e+00
seq_1001_test seq_5209_train 0.95 0.95 96.6 91.67 0.00e+00
seq_2869_test seq_19777_train 0.99 0.98 93.2 91.62 0.00e+00
seq_1721_test seq_25261_train 0.98 0.98 93.4 91.25 0.00e+00
seq_2375_test seq_10235_train 0.95 0.95 96.2 91.10 0.00e+00
seq_990_test seq_22435_train 0.95 0.95 96.0 91.01 0.00e+00
seq_1590_test seq_7109_train 0.94 0.95 96.2 90.91 0.00e+00
seq_1590_test seq_8335_train 0.94 0.95 96.2 90.91 0.00e+00
seq_294_test seq_8633_train 0.94 0.94 96.9 90.89 0.00e+00
seq_534_test seq_3319_train 0.93 0.93 97.3 90.78 0.00e+00
seq_1927_test seq_17438_train 0.98 0.98 92.4 90.55 0.00e+00
seq_1590_test seq_1648_train 0.94 0.94 96.7 90.51 0.00e+00
seq_1900_test seq_19738_train 0.99 0.99 91.2 90.47 0.00e+00
seq_2133_test seq_19738_train 0.99 0.99 91.2 90.47 0.00e+00
seq_1927_test seq_25021_train 0.98 0.99 92.2 90.45 0.00e+00
seq_174_test seq_8633_train 0.93 0.93 97.1 90.40 0.00e+00
seq_746_test seq_14173_train 0.96 0.96 94.3 90.34 0.00e+00
seq_730_test seq_3319_train 0.94 0.95 95.5 90.25 0.00e+00
seq_681_test seq_23769_train 0.96 0.96 94.4 90.25 0.00e+00
seq_356_test seq_10868_train 0.93 0.93 96.9 90.12 0.00e+00
seq_369_test seq_17996_train 0.95 0.95 95.1 90.06 0.00e+00