Datasets:

songlab
/

TraitGym

chrom stringclasses 22 values	pos int64 99.5k 249M	ref stringclasses 4 values	alt stringclasses 4 values	pip float64 0 1	trait stringclasses 259 values	label bool 2 classes	maf float64 0 0.5	ld_score float64 1.53 628	consequence stringclasses 16 values	tss_dist int64 0 1.58M	match_group stringlengths 5 37
1	867,476	C	T	0.00156		false	0.079465	44.053	non_coding_transcript_exon_variant	56,446	non_coding_transcript_exon_variant_0
1	868,052	T	C	0.001791		false	0.077747	44.057	non_coding_transcript_exon_variant	55,870	non_coding_transcript_exon_variant_0
1	868,635	A	G	0.004349		false	0.075255	43.639	non_coding_transcript_exon_variant	55,287	non_coding_transcript_exon_variant_0
1	870,176	T	A	0		false	0.084371	37.271	non_coding_transcript_exon_variant	53,746	non_coding_transcript_exon_variant_0
1	1,052,930	A	G	0.001467		false	0.058385	46.907	non_coding_transcript_exon_variant	18,823	non_coding_transcript_exon_variant_1
1	1,359,031	C	T	0.003126		false	0.025499	89.382	PLS_flank	236	PLS_flank_0
1	1,704,180	T	C	0		false	0.40899	97.129	non_coding_transcript_exon_variant	11,384	non_coding_transcript_exon_variant_2
1	1,989,803	A	G	0		false	0.26523	45.391	dELS	1,125	dELS_3
1	2,293,397	G	A	0.999932	Height	true	0.37057	32.302	dELS	65,077	dELS_0
1	2,299,360	T	C	0.002934		false	0.3705	36.621	dELS	71,040	dELS_0
1	2,315,482	C	G	0		false	0.010493	14.476	non_coding_transcript_exon_variant	72,854	non_coding_transcript_exon_variant_6
1	2,392,128	T	C	0.008659		false	0.44831	48.579	PLS	237	PLS_4
1	2,409,065	C	T	0.003446		false	0.31807	36.147	dELS	3,474	dELS_7
1	2,416,601	T	C	0.001064		false	0.33711	36.202	dELS	2,803	dELS_7
1	2,869,701	G	T	0		false	0.001792	4.8399	DNase-H3K4me3	68,007	DNase-H3K4me3_0
1	2,878,207	G	C	0		false	0.001781	3.6632	DNase-H3K4me3	76,513	DNase-H3K4me3_0
1	3,079,632	C	G	0		false	0.49644	35.326	dELS	10,420	dELS_9
1	3,080,038	T	C	0.999895	MCH,MCV,Plt,RBC	true	0.23272	31.606	dELS	10,826	dELS_1
1	3,409,530	C	G	0.000537		false	0.23433	33.908	dELS	13,038	dELS_1
1	3,409,624	A	G	0		false	0.23428	33.922	dELS	13,132	dELS_1
1	3,431,704	G	C	0.000904		false	0.2206	18.586	dELS	22,960	dELS_6
1	3,530,431	C	T	0.005246		false	0.17759	35.05	pELS	991	pELS_4
1	3,543,648	G	T	0		false	0.11245	14.434	dELS_flank	12,199	dELS_flank_9
1	3,658,759	T	C	0		false	0.11229	17.07	dELS_flank	5,997	dELS_flank_9
1	3,690,596	G	A	0		false	0.066949	26.88	PLS	75	PLS_2
1	3,718,901	G	C	0		false	0.48348	35.17	dELS	20,842	dELS_9
1	3,774,964	A	G	0.999973	Hb,HbA1c,MCHC,RBC	true	0.23057	95.317	dELS	2,138	dELS_2
1	3,840,004	C	T	0		false	0.001743	6.7618	intron_variant	17,189	intron_variant_2
1	3,904,770	G	A	0		false	0.27154	47.968	dELS	4,497	dELS_3
1	3,959,637	A	G	0		false	0.37341	27.032	dELS	59,364	dELS_0
1	4,571,025	C	G	0		false	0.077068	84.633	pELS	83,583	pELS_3
1	4,694,484	C	T	0		false	0.34943	25.535	dELS_flank	39,751	dELS_flank_5
1	4,725,471	C	T	0		false	0.006133	7.7017	DNase-H3K4me3	70,738	DNase-H3K4me3_0
1	5,460,017	C	G	0		false	0.46288	37.6	dELS	530,909	dELS_12
1	5,486,106	T	C	0		false	0.35941	42.571	dELS	504,820	dELS_12
1	6,058,884	T	C	0		false	0.22973	92.037	dELS	6,868	dELS_2
1	6,100,503	A	G	0		false	0.21939	48.859	3_prime_UTR_variant	6,268	3_prime_UTR_variant_1
1	6,100,638	A	G	0		false	0.21997	48.887	3_prime_UTR_variant	6,133	3_prime_UTR_variant_1
1	6,100,816	A	G	0		false	0.21939	48.836	3_prime_UTR_variant	5,955	3_prime_UTR_variant_1
1	6,100,898	A	G	0		false	0.2197	48.831	3_prime_UTR_variant	5,873	3_prime_UTR_variant_1
1	6,104,620	C	A	0		false	0.063585	21.243	3_prime_UTR_variant	2,151	3_prime_UTR_variant_3
1	6,104,626	C	A	0		false	0.06176	22.678	3_prime_UTR_variant	2,145	3_prime_UTR_variant_3
1	6,236,178	T	G	0.001313		false	0.44911	43.011	PLS	213	PLS_4
1	6,247,570	G	A	0		false	0.043267	14.449	3_prime_UTR_variant	3,390	3_prime_UTR_variant_3
1	6,524,747	T	A	0		false	0.099925	65.263	3_prime_UTR_variant	4,672	3_prime_UTR_variant_2
1	6,812,085	C	T	0		false	0.008041	16.361	dELS_flank	26,553	dELS_flank_13
1	6,892,067	T	G	0		false	0.15268	121.5	dELS	106,535	dELS_17
1	7,468,786	C	T	0.000526		false	0.27177	36.888	dELS_flank	195,498	dELS_flank_12
1	7,469,958	T	G	0.000619		false	0.2947	39.842	dELS_flank	194,326	dELS_flank_12
1	7,472,230	C	T	0.001635		false	0.26527	37.012	dELS_flank	192,054	dELS_flank_12
1	7,506,006	G	A	0		false	0.27659	31.236	dELS	158,278	dELS_27
1	7,509,847	C	A	0		false	0.27276	30.199	dELS	154,437	dELS_27
1	7,509,946	C	A	0		false	0.27727	30.38	dELS	154,338	dELS_27
1	7,510,343	T	C	0		false	0.26775	27.871	dELS	153,941	dELS_27
1	7,905,951	C	T	0		false	0.001823	6.9218	intergenic_variant	29,148	intergenic_variant_3
1	8,941,902	T	A	0		false	0.14329	56.84	dELS	3,964	dELS_18
1	8,945,251	C	T	0.000528		false	0.01779	14.655	pELS	615	pELS_1
1	9,042,898	C	T	0		false	0.38311	86.804	dELS	16,474	dELS_10
1	9,062,449	G	A	0		false	0.35391	82.624	dELS_flank	7,154	dELS_flank_8
1	9,180,545	C	T	0.000542		false	0.060061	35.611	non_coding_transcript_exon_variant	51,442	non_coding_transcript_exon_variant_0
1	9,181,780	G	A	1	AST,Mono	true	0.074322	35.472	non_coding_transcript_exon_variant	52,677	non_coding_transcript_exon_variant_0
1	9,295,877	G	T	0.993319	DVT	true	0.26506	46.307	dELS	1,348	dELS_3
1	9,299,718	A	T	0.005775		false	0.2667	48.34	dELS	5,189	dELS_3
1	9,315,952	G	A	0.002646		false	0.45496	39.813	dELS	21,423	dELS_21
1	9,362,918	T	C	0.000513		false	0.23307	34.994	dELS	7,210	dELS_1
1	9,364,340	A	G	0		false	0.22994	35.367	dELS	8,632	dELS_1
1	9,433,231	C	T	0.00839		false	0.073274	45.728	intron_variant	77,523	intron_variant_10
1	9,439,187	G	A	0		false	0.17008	43.092	pELS	83,479	pELS_3
1	9,585,043	C	T	0		false	0.054223	19.606	3_prime_UTR_variant	3,867	3_prime_UTR_variant_3
1	9,692,590	C	T	0		false	0.41698	65.763	intron_variant	3,019	intron_variant_1
1	10,211,630	C	G	0.999686	Plt	true	0.11408	123.87	5_prime_UTR_variant	13	5_prime_UTR_variant_0
1	10,403,534	T	G	0.000977		false	0.026385	82.276	dELS_flank	3,901	dELS_flank_1
1	10,630,891	C	T	0.985187	eGFR	true	0.059295	52.934	dELS	14,184	dELS_4
1	10,692,299	T	C	0.99998	eGFRcys	true	0.2895	38.555	dELS_flank	47,222	dELS_flank_0
1	10,729,577	C	T	0.002886		false	0.36561	37.956	dELS	67,068	dELS_0
1	10,729,640	T	C	0.001415		false	0.36326	37.842	dELS	67,005	dELS_0
1	10,789,368	G	C	0.007112		false	0.22561	32.86	dELS	7,277	dELS_1
1	10,805,966	C	T	0		false	0.17489	17.134	intergenic_variant	9,315	intergenic_variant_8
1	10,805,973	A	T	0		false	0.17584	17.108	intergenic_variant	9,322	intergenic_variant_8
1	10,810,992	A	C	0		false	0.1796	17.659	intergenic_variant	14,341	intergenic_variant_8
1	10,994,429	C	A	0		false	0.1753	19.471	intergenic_variant	12,352	intergenic_variant_8
1	11,651,348	G	A	0		false	0.11644	12.609	dELS_flank	3,026	dELS_flank_9
1	11,664,035	G	C	0.000678		false	0.47269	39.848	PLS	164	PLS_4
1	11,673,018	T	C	0		false	0.2303	38.945	non_coding_transcript_exon_variant	4,321	non_coding_transcript_exon_variant_4
1	11,848,089	C	G	0.999999	MAP,SBP	true	0.010641	47.372	5_prime_UTR_variant	255	5_prime_UTR_variant_1
1	11,876,856	C	T	0.000721		false	0.41386	131.81	pELS	17,910	pELS_6
1	11,877,306	C	T	0.0005		false	0.40983	134.9	pELS	18,360	pELS_6
1	12,039,288	A	G	1	Eosino	true	0.099751	53.631	dELS	19,789	dELS_5
1	12,040,885	A	T	0		false	0.088348	67.062	dELS	21,386	dELS_14
1	12,044,283	G	C	0		false	0.088431	66.75	dELS	19,019	dELS_14
1	12,150,629	A	G	0.000784		false	0.48917	37.025	dELS	16,361	dELS_9
1	12,158,564	G	A	0		false	0.48127	34.973	dELS	8,426	dELS_9
1	12,186,118	C	T	0.999987	Eosino	true	0.2179	17.311	dELS	19,114	dELS_6
1	12,218,875	A	G	0		false	0.22412	16.868	dELS	11,154	dELS_6
1	12,601,409	G	C	0.914738	RBC	true	0.32957	37.798	dELS	5,100	dELS_7
1	12,615,695	G	A	0		false	0.32085	36.525	dELS	1,099	dELS_7
1	12,617,607	G	A	0		false	0.009083	6.3441	5_prime_UTR_variant	602	5_prime_UTR_variant_3
1	12,618,735	C	T	0		false	0.22592	21.571	non_coding_transcript_exon_variant	524	non_coding_transcript_exon_variant_4
1	12,758,908	A	G	0		false	0.060271	51.889	dELS	12,482	dELS_4
1	13,832,260	T	G	0		false	0.37317	36.572	dELS	60,531	dELS_0

End of preview. Expand in Dataset Viewer.

TraitGym

Benchmarking DNA Sequence Models for Causal Regulatory Variant Prediction in Human Genetics

Quick start

Load a dataset

from datasets import load_dataset

dataset = load_dataset("songlab/TraitGym", "mendelian_traits", split="test")

Example notebook to run variant effect prediction with a gLM, runs in 5 min on Google Colab: TraitGym.ipynb

Resources (https://huggingface.co./datasets/songlab/TraitGym)

Datasets: {dataset}/test.parquet
Subsets: {dataset}/subset/{subset}.parquet
Features: {dataset}/features/{features}.parquet
Predictions: {dataset}/preds/{subset}/{model}.parquet
Metrics: {dataset}/{metric}/{subset}/{model}.csv

dataset examples (load_dataset config name):

mendelian_traits_matched_9 (mendelian_traits)
complex_traits_matched_9 (complex_traits)
mendelian_traits_all (mendelian_traits_full)
complex_traits_all (complex_traits_full)

subset examples:

all (default)
3_prime_UTR_variant
disease
BMI

features examples:

GPN-MSA_LLR
GPN-MSA_InnerProducts
Borzoi_L2

model examples:

GPN-MSA_LLR.minus.score
GPN-MSA.LogisticRegression.chrom
CADD+GPN-MSA+Borzoi.LogisticRegression.chrom

metric examples:

AUPRC_by_chrom_weighted_average (main metric)
AUPRC

Code (https://github.com/songlab-cal/TraitGym)

Tries to follow recommended Snakemake structure
GPN-Promoter code is in the main GPN repo

Downloads last month: 42

Size of the auto-converted Parquet files:

258 MB

Number of rows:

16,735,153

Collection including songlab/TraitGym

TraitGym

Collection

3 items • Updated 8 days ago