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 human_ocr_ensembl_test.csv.gz against train set human_ocr_ensembl_train.csv.gz
Data Leakage:!WarningHow a flag is decided →
gb-qc v1.0.0 · run 2026-09-15 22:31:16

Basic Descriptive Statistics

Filename human_ocr_ensembl_test.csv.gz human_ocr_ensembl_train.csv.gz
Number of sequences 34952 139804
Minimum length 84 71
Mean length 326.89 326.21
Maximum length 593 593
!

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 human_ocr_ensembl_test.csv.gz human_ocr_ensembl_train.csv.gz
Data Leakage (percentage) 1.16% 0.49%
Data Leakage (count) 405 691
Similarity Histogram
1117 high-similarity alignments (first 100 shown)
Query (Q) Target (T) Q Cov. T Cov. % Identity % Similarity E-value Alignment
seq_32212_test seq_115112_train 0.99 0.98 100.0 98.40 3.89e-236
seq_14009_test seq_71187_train 0.98 1.00 100.0 98.40 3.89e-137
seq_20511_test seq_132013_train 0.99 1.00 99.5 98.31 4.52e-127
seq_20655_test seq_135058_train 1.00 0.99 98.5 97.71 3.02e-254
seq_11425_test seq_31345_train 0.98 0.99 99.0 97.32 1.83e-276
seq_33112_test seq_133020_train 1.00 1.00 97.7 97.31 4.24e-134
seq_20004_test seq_86197_train 0.98 1.00 99.6 97.21 2.04e-150
seq_30169_test seq_117927_train 0.99 1.00 98.2 97.12 1.00e-179
seq_29049_test seq_114612_train 1.00 0.98 99.1 97.02 1.75e-119
seq_11304_test seq_38820_train 1.00 0.99 98.0 96.92 2.90e-134
seq_34676_test seq_133243_train 0.99 0.99 97.9 96.92 1.92e-95
seq_2896_test seq_33751_train 0.99 0.98 98.9 96.82 1.18e-144
seq_2654_test seq_966_train 0.98 0.97 99.7 96.81 2.14e-234
seq_32487_test seq_95446_train 1.00 1.00 97.1 96.81 1.62e-157
seq_24368_test seq_117634_train 1.00 1.00 96.7 96.70 1.19e-116
seq_33112_test seq_102962_train 0.99 0.99 97.7 96.63 8.97e-133
seq_17606_test seq_85810_train 0.99 0.99 97.7 96.63 2.64e-132
seq_21936_test seq_132013_train 0.97 0.99 99.5 96.61 4.40e-126
seq_32432_test seq_80499_train 0.97 0.97 99.7 96.61 2.40e-187
seq_14550_test seq_23504_train 0.98 1.00 98.2 96.53 1.01e-115
seq_20733_test seq_76086_train 0.99 1.00 97.1 96.52 2.36e-84
seq_22521_test seq_130135_train 1.00 0.99 97.6 96.43 1.30e-171
seq_22542_test seq_107626_train 0.99 0.99 97.3 96.42 3.67e-173
seq_17606_test seq_127852_train 1.00 0.98 98.1 96.33 5.81e-136
seq_20511_test seq_102011_train 1.00 0.98 98.7 96.33 1.00e-125
seq_20989_test seq_74225_train 0.98 1.00 98.3 96.24 4.73e-120
seq_4925_test seq_46802_train 0.98 0.98 98.6 96.23 7.07e-188
seq_17606_test seq_118810_train 0.99 1.00 97.3 96.23 1.19e-130
seq_27319_test seq_118005_train 1.00 0.99 97.0 96.22 3.94e-116
seq_1236_test seq_30507_train 0.98 0.99 98.1 96.14 2.23e-224
seq_19120_test seq_132079_train 0.98 1.00 97.6 96.14 3.01e-128
seq_10134_test seq_5616_train 0.99 0.99 96.7 96.12 1.93e-165
seq_2654_test seq_55041_train 0.97 0.98 99.5 96.12 2.14e-229
seq_31680_test seq_104919_train 1.00 0.99 96.7 96.02 1.96e-298
seq_7681_test seq_48940_train 0.98 0.98 98.3 95.94 2.48e-259
seq_21936_test seq_102011_train 0.98 0.97 98.7 95.94 9.57e-125
seq_33867_test seq_70353_train 0.99 1.00 97.0 95.93 4.31e-134
seq_23468_test seq_131208_train 1.00 1.00 95.9 95.90 5.23e-80
seq_23762_test seq_109811_train 0.99 1.00 97.0 95.84 1.22e-117
seq_4819_test seq_67846_train 0.98 0.97 99.0 95.83 9.07e-174
seq_34512_test seq_121514_train 0.98 0.99 97.5 95.75 1.60e-161
seq_19711_test seq_94516_train 0.98 1.00 97.2 95.74 2.07e-126
seq_21667_test seq_83249_train 0.99 1.00 97.1 95.74 9.47e-102
seq_29013_test seq_102802_train 0.99 1.00 96.5 95.73 1.16e-125
seq_21667_test seq_136828_train 1.00 1.00 95.7 95.70 8.66e-99
seq_16052_test seq_26529_train 0.97 1.00 98.5 95.64 8.80e-284
seq_30430_test seq_94857_train 0.98 0.99 97.1 95.64 2.09e-159
seq_1461_test seq_62931_train 0.99 0.99 96.6 95.63 1.65e-145
seq_22012_test seq_79355_train 0.96 0.99 100.0 95.60 1.80e-151
seq_27012_test seq_646_train 0.98 0.98 97.7 95.55 2.07e-135
seq_24672_test seq_89967_train 0.98 1.00 97.0 95.55 7.21e-133
seq_22521_test seq_74740_train 0.98 1.00 97.2 95.45 1.16e-166
seq_24794_test seq_128164_train 1.00 0.97 97.9 95.35 2.86e-151
seq_33179_test seq_74555_train 0.96 0.99 98.8 95.34 5.97e-189
seq_30862_test seq_74798_train 0.99 1.00 96.4 95.34 2.99e-177
seq_32720_test seq_104849_train 0.95 0.97 100.0 95.30 1.04e-186
seq_32346_test seq_119308_train 0.98 1.00 97.4 95.26 2.50e-134
seq_29253_test seq_646_train 1.00 0.97 98.1 95.26 9.93e-135
seq_24794_test seq_131027_train 1.00 0.98 96.9 95.25 1.22e-145
seq_18827_test seq_104760_train 0.99 1.00 96.5 95.25 1.18e-258
seq_26224_test seq_123191_train 1.00 0.99 96.3 95.24 4.39e-173
seq_20005_test seq_112542_train 0.99 0.99 96.3 95.24 2.24e-131
seq_29253_test seq_95326_train 1.00 0.98 97.3 95.16 2.08e-131
seq_24162_test seq_107578_train 0.98 1.00 97.2 95.16 1.07e-67
seq_25012_test seq_109915_train 1.00 0.98 96.8 95.15 2.75e-138
seq_17606_test seq_138238_train 1.00 0.98 96.6 95.15 2.48e-129
seq_29397_test seq_103375_train 1.00 0.96 98.7 95.15 1.35e-124
seq_4694_test seq_56150_train 0.96 0.99 98.9 95.14 5.77e-258
seq_12463_test seq_24276_train 0.99 1.00 96.2 95.14 1.20e-167
seq_19296_test seq_77468_train 0.97 0.96 99.3 95.13 2.65e-172
seq_23762_test seq_127928_train 1.00 0.98 96.6 95.05 2.61e-117
seq_10153_test seq_22230_train 0.96 1.00 99.1 95.04 2.22e-237
seq_8022_test seq_32641_train 0.95 1.00 99.6 95.02 1.15e-153
seq_24368_test seq_132902_train 1.00 1.00 95.0 95.00 4.01e-110
seq_25241_test seq_98021_train 0.98 1.00 96.9 94.96 1.29e-144
seq_25241_test seq_75484_train 0.98 0.99 96.9 94.96 1.29e-144
seq_19120_test seq_117885_train 0.99 1.00 96.1 94.95 2.48e-122
seq_26224_test seq_121466_train 0.99 1.00 96.0 94.94 7.64e-169
seq_23838_test seq_127584_train 0.99 0.96 99.2 94.93 5.67e-200
seq_12782_test seq_1714_train 1.00 0.99 95.5 94.93 2.77e-148
seq_30430_test seq_119646_train 0.98 0.97 97.5 94.87 4.74e-161
seq_32346_test seq_70322_train 0.98 0.97 97.7 94.87 1.23e-136
seq_22542_test seq_121466_train 0.99 0.98 96.8 94.86 7.13e-170
seq_33639_test seq_119904_train 0.98 0.97 98.1 94.86 2.16e-258
seq_18605_test seq_135388_train 0.98 1.00 96.6 94.86 1.43e-130
seq_716_test seq_7420_train 0.96 0.96 98.8 94.85 1.38e-173
seq_23838_test seq_101321_train 0.99 1.00 95.9 94.85 7.71e-179
seq_272_test seq_1165_train 0.96 0.96 99.1 94.84 1.28e-188
seq_21361_test seq_75750_train 1.00 1.00 94.8 94.80 8.11e-132
seq_4594_test seq_64162_train 0.99 0.97 97.3 94.77 1.16e-152
seq_33112_test seq_119745_train 0.97 1.00 97.7 94.77 1.78e-129
seq_5210_test seq_10766_train 0.97 1.00 97.8 94.77 8.38e-190
seq_19153_test seq_93149_train 0.99 0.96 98.6 94.75 1.64e-73
seq_29049_test seq_118322_train 1.00 0.99 96.0 94.75 3.50e-107
seq_34346_test seq_84570_train 0.95 1.00 99.8 94.71 2.73e-272
seq_31838_test seq_91440_train 0.98 0.97 97.4 94.67 1.01e-137
seq_22521_test seq_126823_train 0.98 0.97 97.8 94.67 2.69e-169
seq_25012_test seq_91440_train 0.98 0.98 96.4 94.66 2.47e-135
seq_19120_test seq_129883_train 0.99 0.98 96.1 94.66 5.52e-123
seq_29636_test seq_101449_train 1.00 1.00 94.6 94.60 1.38e-83