Datasets:

katielink
/

genomic-benchmarks

id int64 0 50k	region stringlengths 4 15	start int64 0 249M	end int64 200 249M	strand stringclasses 1 value
27,434	ENST00000217185	203	403	+
13,400	ENST00000367051	4,701	4,901	+
883	ENST00000360507	623	823	+
7,303	ENST00000489432	1,469	1,669	+
45,124	ENST00000309668	64	264	+
874	ENST00000335754	416	616	+
45,943	ENST00000595541	921	1,121	+
11,458	ENST00000639499	457	657	+
23,877	ENST00000300061	831	1,031	+
11	ENST00000526014	43	243	+
27,813	ENST00000588783	101	301	+
10,713	ENST00000281932	113	313	+
46,014	ENST00000510693	93	293	+
3,645	ENST00000302555	1,363	1,563	+
10,448	ENST00000374358	88	288	+
41,968	ENST00000323829	280	480	+
3,813	ENST00000368206	167	367	+
4,523	ENST00000430285	8	208	+
7,231	ENST00000560550	18	218	+
39,423	ENST00000506192	146	346	+
46,647	ENST00000419602	302	502	+
12,547	ENST00000652508	1	201	+
10,917	ENST00000367992	4	204	+
16,904	ENST00000253458	1,587	1,787	+
39,863	ENST00000626926	42	242	+
24,728	ENST00000329003	784	984	+
12,695	ENST00000395305	210	410	+
36,114	ENST00000374479	491	691	+
29,694	ENST00000413728	1,186	1,386	+
7,528	ENST00000514798	1,156	1,356	+
28,808	ENST00000549706	985	1,185	+
4,756	ENST00000614640	40	240	+
30,432	ENST00000376452	275	475	+
1,108	ENST00000650697	1,417	1,617	+
11,090	ENST00000324907	1,157	1,357	+
42,876	ENST00000565186	659	859	+
28,136	ENST00000542824	427	627	+
19,860	ENST00000424325	206	406	+
8,054	ENST00000265293	521	721	+
5,886	ENST00000543336	1,065	1,265	+
44,551	ENST00000367614	2,356	2,556	+
3,542	ENST00000648665	2,217	2,417	+
8,359	ENST00000409476	234	434	+
15,074	ENST00000392799	751	951	+
40,589	ENST00000332118	965	1,165	+
19,528	ENST00000528880	267	467	+
2,333	ENST00000360839	6,904	7,104	+
37,768	ENST00000665583	16	216	+
37,791	ENST00000539416	14	214	+
22,894	ENST00000367725	298	498	+
11,790	ENST00000451385	290	490	+
43,628	ENST00000592558	144	344	+
47,118	ENST00000518817	73	273	+
17,703	ENST00000649186	468	668	+
14,474	ENST00000517399	779	979	+
22,427	ENST00000524107	34	234	+
28,883	ENST00000443137	842	1,042	+
5,785	ENST00000256854	471	671	+
27,395	ENST00000527838	213	413	+
5,883	ENST00000368757	1	201	+
37,911	ENST00000443584	1,477	1,677	+
7,002	ENST00000389645	1,142	1,342	+
8,801	ENST00000521333	11	211	+
39,803	ENST00000639767	134	334	+
46,757	ENST00000537098	406	606	+
34,684	ENST00000622645	97	297	+
17,117	ENST00000545204	3	203	+
31,532	ENST00000360184	11,384	11,584	+
268	ENST00000450673	305	505	+
3,142	ENST00000532330	410	610	+
4,800	ENST00000627748	1	201	+
35,868	ENST00000651178	575	775	+
7,422	ENST00000598409	4	204	+
21,413	ENST00000460465	524	724	+
4,396	ENST00000357947	1,110	1,310	+
2,156	ENST00000370591	1,655	1,855	+
28,372	ENST00000566658	2	202	+
32,050	ENST00000412504	641	841	+
1,441	ENST00000295685	83	283	+
25,015	ENST00000527142	238	438	+
25,072	ENST00000368471	2,247	2,447	+
27,121	ENST00000524945	718	918	+
49,882	ENST00000340446	1,850	2,050	+
19,546	ENST00000359359	469	669	+
35,893	ENST00000650547	58	258	+
13,905	ENST00000646968	9	209	+
32,947	ENST00000383170	20	220	+
15,558	ENST00000466197	184	384	+
6,527	ENST00000389568	260	460	+
49,722	ENST00000382821	131	331	+
17,487	ENST00000450845	409	609	+
37,634	ENST00000584288	1	201	+
16,966	ENST00000481785	1,365	1,565	+
12,858	ENST00000622072	472	672	+
48,444	ENST00000373955	2,629	2,829	+
11,146	ENST00000317827	394	594	+
24,050	ENST00000394976	23	223	+
1,496	ENST00000340645	412	612	+
733	ENST00000343820	103	303	+
44,488	ENST00000652479	361	561	+

End of preview. Expand in Dataset Viewer.

Genomic Benchmark

In this repository, we collect benchmarks for classification of genomic sequences. It is shipped as a Python package, together with functions helping to download & manipulate datasets and train NN models.

Citing Genomic Benchmarks

If you use Genomic Benchmarks in your research, please cite it as follows.

Text

GRESOVA, Katarina, et al. Genomic Benchmarks: A Collection of Datasets for Genomic Sequence Classification. bioRxiv, 2022.

BibTeX

@article{gresova2022genomic,
  title={Genomic Benchmarks: A Collection of Datasets for Genomic Sequence Classification},
  author={Gresova, Katarina and Martinek, Vlastimil and Cechak, David and Simecek, Petr and Alexiou, Panagiotis},
  journal={bioRxiv},
  year={2022},
  publisher={Cold Spring Harbor Laboratory},
  url={https://www.biorxiv.org/content/10.1101/2022.06.08.495248}
}

From the github repo:

Datasets

Each folder contains either one benchmark or a set of benchmarks. See docs/ for code used to create these benchmarks.

Naming conventions

dummy_...: small datasets, used for testing purposes
demo_...: middle size datasets, not necesarily biologically relevant or fully reproducible, used in demos

Versioning

We recommend to check the version number when working with the dataset (i.e. not using default None). The version should be set to 0 when the dataset is proposed, after inicial curration it should be changed to 1 and then increased after every modification.

Data format

Each benchmark should contain metadata.yaml file with its main folder with the specification in YAML format, namely

the version of the benchmark (0 = in development)
the classes of genomic sequences, for each class we further need to specify
- url with the reference
- type of the reference (currently, only fa.gz implemented)
- extra_processing, a parameter helping to overcome some know issues with identifiers matching

The main folder should also contain two folders, train and test. Both those folders should contain gzipped CSV files, one for each class (named class_name.csv.gz).

The format of gzipped CSV files closely resemble BED format, the column names must be the following:

id: id of a sequence
region: chromosome/transcript/... to be matched with the reference
start, end: genomic interval specification (0-based, i.e. same as in Python)
strand: either '+' or '-'

To contribute a new datasets

Create a new branch. Add the new subfolders to datasets and docs. The subfolder of docs should contain a description of the dataset in README.md. If the dataset comes with the paper, link the paper. If the dataset is not taken from the paper, make sure you have described and understand the biological process behind it.

If you have access to cloud_cache folder on GDrive, upload your file there and update CLOUD_CACHE in cloud_caching.py.

To review a new dataset

Make sure you can run and reproduce the code. Check you can download the actual sequences and/or create a data loader. Do you understand what is behind these data? (either from the paper or the description) Ask for clarification if needed.

Downloads last month: 368

Size of downloaded dataset files:

9.69 MB

Size of the auto-converted Parquet files:

18.5 MB

Number of rows:

847,660