Datasets:

DeweyWang
/

ETH_ML4G_Project-1

Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 1 new columns ({'gex'}) and 6 missing columns ({'TSS_start', 'gene_end', 'chr', 'strand', 'gene_start', 'TSS_end'}).

This happened while the csv dataset builder was generating data using

hf://datasets/DeweyWang/ETH_ML4G_Project-1/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_train_y.tsv (at revision f983b25b01cde80ee872a5428e0d9886b42239a6), ['hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_train_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_train_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_val_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_val_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_train_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_train_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_val_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_val_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X3_test_info.tsv']

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1890, in _prepare_split_single
                  writer.write_table(table)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 760, in write_table
                  pa_table = table_cast(pa_table, self._schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              gene_name: string
              gex: double
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 488
              to
              {'gene_name': Value('string'), 'chr': Value('string'), 'gene_start': Value('int64'), 'gene_end': Value('int64'), 'TSS_start': Value('int64'), 'TSS_end': Value('int64'), 'strand': Value('string')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1342, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 907, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1739, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1892, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 1 new columns ({'gex'}) and 6 missing columns ({'TSS_start', 'gene_end', 'chr', 'strand', 'gene_start', 'TSS_end'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/DeweyWang/ETH_ML4G_Project-1/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_train_y.tsv (at revision f983b25b01cde80ee872a5428e0d9886b42239a6), ['hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_train_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_train_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_val_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X1_val_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_train_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_train_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_val_info.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X2_val_y.tsv', 'hf://datasets/DeweyWang/ETH_ML4G_Project-1@f983b25b01cde80ee872a5428e0d9886b42239a6/ML4G_Project_1_Data/CAGE-train/CAGE-train/X3_test_info.tsv']
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

gene_name string	chr string	gene_start int64	gene_end int64	TSS_start int64	TSS_end int64	strand string
SLC20A1	chr2	112,645,939	112,663,825	112,658,362	112,658,412	+
C11orf58	chr11	16,613,132	16,758,340	16,738,643	16,738,693	+
ZSCAN9	chr6	28,224,886	28,233,487	28,225,263	28,225,313	+
CD19	chr16	28,931,965	28,939,342	28,931,956	28,932,006	+
TMEM123	chr11	102,396,332	102,470,384	102,452,789	102,452,839	-
POMC	chr2	25,160,853	25,168,903	25,168,640	25,168,690	-
NME2	chr17	51,165,435	51,171,744	51,169,382	51,169,432	+
CEP120	chr5	123,344,890	123,423,592	123,423,542	123,423,592	-
FBXO4	chr5	41,925,254	41,941,743	41,925,253	41,925,303	+
GRIA1	chr5	153,489,615	153,813,869	153,492,171	153,492,221	+
GALNT9	chr12	132,196,372	132,328,920	132,257,692	132,257,742	-
AKAIN1	chr18	5,142,911	5,197,503	5,197,453	5,197,503	-
SYCE1L	chr16	77,199,408	77,213,215	77,199,396	77,199,446	+
GOLGA8S	chr15	23,354,748	23,367,231	23,354,747	23,354,797	+
CAST	chr5	96,525,267	96,779,595	96,702,788	96,702,838	+
FBXW12	chr3	48,372,219	48,401,259	48,379,736	48,379,786	+
DIRAS2	chr9	90,609,832	90,642,862	90,643,054	90,643,104	-
WDR7	chr18	56,651,343	57,029,811	56,890,832	56,890,882	+
MMP1	chr11	102,789,401	102,798,160	102,798,110	102,798,160	-
SREBF2	chr22	41,833,079	41,907,307	41,885,595	41,885,645	+
SLC38A3	chr3	50,205,246	50,221,486	50,205,267	50,205,317	+
PCDHGA11	chr5	141,421,047	141,512,975	141,421,046	141,421,096	+
ZNF707	chr8	143,684,452	143,713,898	143,684,461	143,684,511	+
ACTR6	chr12	100,199,122	100,241,865	100,199,121	100,199,171	+
POLR3K	chr16	46,407	53,608	52,092	52,142	-
POLR2F	chr22	37,952,607	38,041,915	37,953,662	37,953,712	+
CSDC2	chr22	41,561,010	41,577,741	41,572,016	41,572,066	+
POGLUT3	chr11	108,472,112	108,498,384	108,486,187	108,486,237	-
ASB16	chr17	44,170,447	44,179,084	44,170,705	44,170,755	+
GRIK2	chr6	100,962,701	102,081,622	101,927,778	101,927,828	+
LY6G6E	chr6	31,711,771	31,714,065	31,713,762	31,713,812	-
ADAM2	chr8	39,743,735	39,838,227	39,838,177	39,838,227	-
OR6M1	chr11	123,805,408	123,806,349	123,806,337	123,806,387	-
MCF2L2	chr3	183,178,041	183,428,778	183,180,263	183,180,313	-
C3orf18	chr3	50,558,025	50,571,027	50,567,626	50,567,676	-
CLEC4E	chr12	8,533,305	8,540,905	8,540,855	8,540,905	-
TXN	chr9	110,243,810	110,256,507	110,245,066	110,245,116	-
CD109	chr6	73,695,785	73,828,316	73,792,654	73,792,704	+
CNTNAP3	chr9	39,064,710	39,288,315	39,088,423	39,088,473	-
MAD1L1	chr7	1,815,793	2,233,243	2,230,611	2,230,661	-
CYP51A1	chr7	92,084,987	92,134,803	92,134,480	92,134,530	-
ALK	chr2	29,192,774	29,921,586	29,717,650	29,717,700	-
CABP4	chr11	67,452,406	67,461,752	67,452,418	67,452,468	+
MYF5	chr12	80,716,912	80,719,671	80,716,911	80,716,961	+
NUP42	chr7	23,181,841	23,201,011	23,194,402	23,194,452	+
APEH	chr3	49,674,014	49,683,971	49,674,359	49,674,409	+
TMEM41B	chr11	9,280,654	9,314,636	9,314,583	9,314,633	-
SEC61A2	chr10	12,129,637	12,169,961	12,129,693	12,129,743	+
PDE2A	chr11	72,576,141	72,674,591	72,674,400	72,674,450	-
SI	chr3	164,978,898	165,078,496	165,075,964	165,076,014	-
KIF3B	chr20	32,277,651	32,335,011	32,277,663	32,277,713	+
ARHGDIB	chr12	14,942,031	14,961,728	14,961,216	14,961,266	-
RGS11	chr16	268,301	275,980	275,864	275,914	-
NQO2	chr6	2,987,987	3,019,755	3,000,157	3,000,207	+
APBB1	chr11	6,395,125	6,419,414	6,419,061	6,419,111	-
TBC1D3I	chr17	36,355,285	36,366,216	36,264,503	36,264,553	-
PBK	chr8	27,809,624	27,838,082	27,837,759	27,837,809	-
VPS37A	chr8	17,246,931	17,302,427	17,247,122	17,247,172	+
FAS	chr10	88,953,813	89,029,605	88,990,851	88,990,901	+
GTF3C2	chr2	27,325,849	27,357,034	27,337,258	27,337,308	-
P4HA1	chr10	73,007,217	73,096,974	73,096,800	73,096,850	-
ZNF680	chr7	64,519,878	64,563,075	64,563,025	64,563,075	-
NT5DC1	chr6	116,100,851	116,249,497	116,110,851	116,110,901	+
GPR6	chr6	109,978,256	109,980,720	109,979,112	109,979,162	+
CNNM2	chr10	102,918,294	103,090,222	102,918,292	102,918,342	+
LRP4	chr11	46,856,717	46,918,642	46,889,637	46,889,687	-
RAET1L	chr6	150,018,334	150,025,532	150,025,482	150,025,532	-
RGL4	chr22	23,688,136	23,699,176	23,691,305	23,691,355	+
NUP43	chr6	149,724,315	149,749,665	149,746,502	149,746,552	-
PSMC5	chr17	63,827,152	63,832,026	63,827,432	63,827,482	+
HSPA12B	chr20	3,732,685	3,753,111	3,732,710	3,732,760	+
CSTF2T	chr10	51,695,486	51,699,595	51,699,541	51,699,591	-
SPPL2C	chr17	45,844,881	45,847,067	45,844,834	45,844,884	+
CLPSL1	chr6	35,781,019	35,793,675	35,781,016	35,781,066	+
MRO	chr18	50,795,120	50,825,402	50,819,786	50,819,836	-
TUT1	chr11	62,575,045	62,591,637	62,591,587	62,591,637	-
NHLRC3	chr13	39,038,306	39,050,109	39,038,310	39,038,360	+
TYSND1	chr10	70,137,981	70,146,700	70,146,626	70,146,676	-
QRICH1	chr3	49,029,707	49,094,363	49,093,561	49,093,611	-
AP3S1	chr5	115,841,592	115,914,081	115,895,086	115,895,136	+
BSX	chr11	122,977,570	122,981,834	122,981,670	122,981,720	-
CHD6	chr20	41,402,083	41,618,384	41,499,294	41,499,317	-
B3GALT4	chr6	33,255,053	33,256,746	33,278,687	33,278,737	+
RNF216	chr7	5,620,047	5,781,696	5,649,965	5,650,015	-
LRFN2	chr6	40,391,591	40,587,364	40,587,415	40,587,465	-
CCDC28A	chr6	138,773,769	138,793,319	138,773,768	138,773,818	+
PRPSAP1	chr17	76,309,478	76,384,521	76,384,471	76,384,521	-
OR5T1	chr11	56,275,639	56,276,619	56,275,638	56,275,688	+
PPP1R2B	chr5	156,850,295	156,852,528	156,850,562	156,850,612	+
ADH7	chr4	99,412,261	99,435,510	99,430,241	99,430,291	-
PTPN18	chr2	130,356,045	130,375,405	130,358,949	130,358,975	+
MET	chr7	116,672,196	116,798,377	116,758,516	116,758,566	+
KCNIP2	chr10	101,825,974	101,843,920	101,843,518	101,843,568	-
ENDOD1	chr11	95,089,846	95,132,645	95,089,809	95,089,859	+
TAAR5	chr6	132,588,592	132,589,741	132,589,636	132,589,686	-
EPM2AIP1	chr3	36,985,043	36,993,131	36,992,890	36,992,940	-
MAMDC4	chr9	136,850,943	136,860,799	136,852,366	136,852,416	+
CDK7	chr5	69,234,793	69,277,413	69,234,840	69,234,890	+
IFNA7	chr9	21,201,469	21,202,205	21,202,155	21,202,205	-
CYLC2	chr9	102,995,311	103,018,488	102,995,310	102,995,360	+

End of preview.

Gene Expression Prediction Dataset

📌 Overview

This dataset is designed for predicting gene expression levels from chromatin landscape data, including histone modifications and chromatin accessibility.

It is part of a machine learning project in genomics, where the goal is to model the relationship between epigenetic signals and gene expression.

👉 Full project code (including preprocessing and prediction): https://github.com/Dewey-Wang/Gene-expression-prediction/tree/main

📂 Dataset Structure

The dataset consists of two main components:

1. Raw Data

Total size: 18.66 GB
Number of files: 72
Includes:
- Histone modification data (ChIP-seq)
- Chromatin accessibility (DNase-seq)
- Gene expression (CAGE)
- Gene annotation (TSS, gene body, RefSeq)

2. Preprocessed Data

Total size: 6.36 GB
Number of files: 53
Includes:
- Feature matrices for machine learning
- Aggregated signals around genomic regions (e.g. TSS windows)
- Normalized inputs ready for model training

👉 Full preprocessing code is available in the GitHub repository above.

🎯 Task

The main task is:

Predict gene expression levels from chromatin features

Input: epigenetic signals (ChIP-seq, DNase-seq)
Output: gene expression values

📊 Evaluation

Typical evaluation metrics:

Spearman correlation (primary)
Pearson correlation
R² score

🧬 Data Details

Genome version: hg38 / GRCh38
Multiple cell lines included
Data normalized for cross-cell-line comparison

🚀 Usage

You can either:

Use preprocessed data directly for ML models
Reproduce preprocessing using provided code

⚠️ Notes

Raw data is large (~18.66 GB)
Preprocessed data is recommended for quick experimentation
Suitable for machine learning and bioinformatics research

Downloads last month: -