data/GSBench.jsonl

{"id": 1, "task": "Given a VCF-formatted genotype file located at {Q001/raw.vcf.gz}, perform original data quality control (QC). Filter out variants with a missing rate >10% and a MAF <0.05. Convert the QC-filtered data into PLINK binary format (geno_qc.bed/bim/fam). Calculate the first five principal components (PCs) and save the proportion of variance explained by each PC to pca_results.txt.", "reference_steps": 4, "categories": ["Data quality control", "Population genetic structure analysis"], "data_files": ["files/Q001/raw.vcf.gz"], "reference_answer": ["reference_answer/Q001"]}
{"id": 2, "task": "Given a VCF-formatted genotype file {Q002/filter.vcf.gz}, first convert it to PLINK binary format. Using the converted file filter.bed, extract specified samples based on the provided pedigree information file {Q002/egMater.txt}. Perform LD decay analysis on the extracted subset, restricting calculations to Chr1 positions between 1,000,000 and 2,000,000 bp. Set the sliding window to 50 SNPs and the r² threshold to 0, then calculate the mean LD decay distance.", "reference_steps": 3, "categories": ["Data quality control", "Population genetic structure analysis"], "data_files": ["files/Q002/filter.vcf.gz", "files/Q002/egMater.txt"], "reference_answer": ["reference_answer/Q002"]}
{"id": 3, "task": "Given a Phenotypic data file {Q003/phe.csv} containing 9 environments for the trait “Plant Height”. Firstly, calculate the BLUPs for this trait in each environment. Subsequently, conduct a GWAS using the calculated BLUPs as the phenotypic input, in conjunction with the genotype file {Q003/maize_598.vcf.gz}. Perform GWAS using three methods, specifically FarmCPU, GLM, and MLM, while specifying the parameters with vc.method=“BRENT”, method.bin=“static”, and threshold=0.05.", "reference_steps": 3, "categories": ["Genetic parameter estimation & Genomic prediction", "Gene mining & Functional characterization"], "data_files": ["files/Q003/phe.csv", "files/Q003/maize_598.vcf.gz"], "reference_answer": ["reference_answer/Q003"]}
{"id": 4, "task": "Given a genotype file {Q004/maize_598.vcf.gz} and a phenotype file {Q004/phe.csv} containing 9 environments, perform variance component estimation for G×E using raw phenotypic data. Estimate the proportion of total phenotypic variance explained by genetic (G), environmental (E), and G×E variance components. Finally, calculate the BLUPs for the “Plant Height” trait under each environment.", "reference_steps": 3, "categories": ["Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q004/maize_598.vcf.gz", "files/Q004/phe.csv"], "reference_answer": ["reference_answer/Q004"]}
{"id": 5, "task": "Perform quality control on the VCF file {Q005/raw_snps.vcf}. Filter out variants with a variant missing rate >10%, a sample missing rate >5%, and those failing the Hardy-Weinberg Equilibrium test (P-value < 1e-6). Convert the filtered VCF into an additive genotype count matrix (0,1,2 format) and save the resulting matrix as a txt file.", "reference_steps": 3, "categories": ["Data quality control"], "data_files": ["files/Q005/raw_snps.vcf"], "reference_answer": ["reference_answer/Q005"]}
{"id": 6, "task": "Given a genotype file {Q006/maize_598.vcf.gz}and a phenotype file {Q006/phe.csv} containing data from 9 environments, focus on the “Plant Height” trait. First, calculate the environmental response slope and intercept for each line using a reaction norm model. Then, treat the calculated slope as a new phenotypic trait and perform GWAS using this slope and the genotype data to identify loci controlling environmental adaptation. Execute GWAS using three methods, specifically FarmCPU, GLM, and MLM, while specifying the parameters with vc.method=“BRENT”, method.bin=“static”, and threshold=0.05.", "reference_steps": 4, "categories": ["Environmental & Phenotypic data analysis", "Gene mining & Functional characterization"], "data_files": ["files/Q006/maize_598.vcf.gz", "files/Q006/phe.csv"], "reference_answer": ["reference_answer/Q006"]}
{"id": 7, "task": "Given a genotype file {Q007/filter.vcf.gz}, first perform population genetic diversity assessment by calculating the proportion of polymorphic sites and the average nucleotide diversity (π). Second, based on the given group information {Q007/group.txt}, divide the lines into two groups and compare the allele frequencies between these groups, saving the comparison results as a txt file.", "reference_steps": 4, "categories": ["Population genetic structure analysis"], "data_files": ["files/Q007/filter.vcf.gz", "files/Q007/group.txt"], "reference_answer": ["reference_answer/Q007"]}
{"id": 8, "task": "Given a phenotype file {Q008/cross_performance.csv} containing parental and hybrid phenotypic data, first estimate the GCA for all parents and the SCA for all crosses. Then, identify the top 3 male and top 3 female parents based on their GCA values. Finally, filter the data again to list the 9 hybrid combinations derived from these selected parents along with their SCA values, saving the output to result.txt.", "reference_steps": 4, "categories": ["Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q008/cross_performance.csv"], "reference_answer": ["reference_answer/Q008"]}
{"id": 9, "task": "Given a genotype file {Q009/filter.vcf.gz}, first perform LD pruning with a window size of 50, a step size of 5, and an R² threshold of 0.2 to obtain a PLINK binary file containing an independent SNP subset. Then, perform PCA using the subset and retain the first 10 principal components.", "reference_steps": 3, "categories": ["Data quality control", "Population genetic structure analysis"], "data_files": ["files/Q009/filter.vcf.gz"], "reference_answer": ["reference_answer/Q009"]}
{"id": 10, "task": "Given a gene list {Q010/candidate_genes.txt}, first perform GO and KEGG functional enrichment analyses for the genes in the list. Then, annotate the protein domains of these genes and identify statistically enriched PFAM domains. Apply significance thresholds of P-value < 0.05 for GO enrichment analysis, and P-value < 0.1 coupled with FDR < 0.05 for KEGG pathway analysis.", "reference_steps": 7, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q010/candidate_genes.txt"], "reference_answer": ["reference_answer/Q010"]}
{"id": 11, "task": "Given a population genotype file {Q011/maize_598.vcf.gz}, calculate the GRM among individuals. Given a phenotype file {Q011/phe.csv}, estimate the heritability of the trait in each environment using the phenotypic data and the GRM calculated in the previous step.", "reference_steps": 3, "categories": ["Population genetic structure analysis", "Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q011/maize_598.vcf.gz", "files/Q011/phe.csv"], "reference_answer": ["reference_answer/Q011"]}
{"id": 12, "task": "Perform genotype imputation on {Q012/Chip4w.vcf.gz}to reduce the missing rate. Compare the distribution of variant missing rates before and after imputation, and save the results to a file.", "reference_steps": 3, "categories": ["Data quality control"], "data_files": ["files/Q012/Chip4w.vcf.gz"], "reference_answer": ["reference_answer/Q012"]}
{"id": 13, "task": "Given environmental factor data {Q013/5Envs_envParas_DAP150.txt} and planting information {Q013/Env_meta_table.txt}, first calculate the correlations among phenotypic traits across different environments and generate a heatmap. Then, calculate the environmental index (EI) and use the CERIS method to identify the environmental factors and critical growth windows that exhibit the highest correlation with EI.", "reference_steps": 5, "categories": ["Environmental & Phenotypic data analysis"], "data_files": ["files/Q013/5Envs_envParas_DAP150.txt", "files/Q013/Env_meta_table.txt"], "reference_answer": ["reference_answer/Q013"]}
{"id": 14, "task": "Given a population genotype file {Q014/maize_598.vcf.gz}and a phenotype file {Q014/phe.csv}, first perform GWAS. Execute GWAS using three methods, specifically FarmCPU, GLM, and MLM, while specifying the parameters with vc.method=“BRENT”, method.bin=“static”, and threshold=0.05. Save the significant loci identified by the FarmCPU method as a list in loci.txt, with each line representing one locus. Subsequently, extract these loci from {Q014/maize_598.vcf.gz} based on loci.txt to generate a new genotype file new.vcf. Finally, calculate the allele frequencies for new.vcf.", "reference_steps": 5, "categories": ["Gene mining & Functional characterization", "Data quality control"], "data_files": ["files/Q014/maize_598.vcf.gz", "files/Q014/phe.csv", "files/Q014/maize_598.vcf.gz"], "reference_answer": ["reference_answer/Q014"]}
{"id": 15, "task": "Given a list of significant GWAS loci {Q015/locilist.txt} mapped to the B73 V4 reference genome, define flanking regions by extending 50 kb upstream and downstream of each locus. Extract all genes residing within these expanded intervals. Subsequently, retrieve orthologous genes for these candidates from Arabidopsis thaliana and Oryza sativa and export the comparative results into gene.txt.", "reference_steps": 3, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q015/locilist.txt"], "reference_answer": ["reference_answer/Q015"]}
{"id": 16, "task": "Given a genotype file located at {Q016/geno.vcf}, remove sites with a missing rate greater than 5%, and then convert the VCF genotype data into a CSV genotype file named geno.csv that can be used for genomic prediction. Next, given a phenotype file located at {Q016/phe.csv}, perform z-score normalization on PH_JL and save the processed phenotype data as new_phe.csv. Generate a cross-validation file named cvf.csv based on the material names in new_phe.csv. Finally, use geno.csv, new_phe.csv, and cvf.csv to perform genomic prediction with the rrBLUP method, using the z-score-normalized PH_JL as the target phenotype and 10-fold cross-validation for model evaluation.", "reference_steps": 6, "categories": ["Data quality control", "Genetic parameter estimation & Genomic prediction modeling"], "data_files": ["files/Q016/geno.vcf", "files/Q016/phe.csv"], "reference_answer": ["reference_answer/Q016"]}
{"id": 17, "task": "First, perform genotype imputation on the file {Q017/Chip4w.vcf.gz}. Given the multi-environment phenotype file {Q017/phe.csv}, reformat it into new.txt to meet the input requirements for heritability estimation. Then filter new.txt to retain only the samples present in the genotype file. Finally, estimate the heritability of the trait across multiple environments.", "reference_steps": 4, "categories": ["Data quality control", "Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q017/Chip4w.vcf.gz", "files/Q017/phe.csv"], "reference_answer": ["reference_answer/Q017"]}
{"id": 18, "task": "Using the genotype file {Q018/maize_598.vcf.gz}and the phenotype file {Q018/phe.csv}, execute GWAS. Implement three models, specifically FarmCPU, GLM, and MLM, while specifying the parameters with vc.method=“BRENT”, method.bin=“static”, and threshold=0.05. Extract the significant loci identified by the FarmCPU model and save them to loci.txt.", "reference_steps": 3, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q018/maize_598.vcf.gz", "files/Q018/phe.csv"], "reference_answer": ["reference_answer/Q018"]}
{"id": 19, "task": "Given a phenotype file {Q019/phe.csv} containing “Plant Height” data across multiple environments and a genotype file {Q019/geno.vcf}, first convert the genotype file into a CSV matrix suitable for genomic prediction. Then, calculate the BLUPs for each variety across all environments and save these values as phe.csv. Subsequently, extract the sample names from phe.csv and save them to name.txt. Using name.txt, generate a ten-fold cross-validation grouping file and save it as cvf.csv. Finally, perform genomic prediction using the rrBLUP model with ten-fold cross-validation and report the Pearson correlation coefficient between the predicted and observed BLUP values.", "reference_steps": 5, "categories": ["Genetic parameter estimation & Genomic prediction modeling"], "data_files": ["files/Q019/phe.csv", "files/Q019/geno.vcf"], "reference_answer": ["reference_answer/Q019"]}
{"id": 20, "task": "Perform PFAM annotation for the sequences corresponding to the gene list {Q020/candidate_genes.txt}. Following annotation, conduct PFAM domain enrichment analysis on the results, using a significance threshold of FDR < 0.05.", "reference_steps": 4, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q020/candidate_genes.txt"], "reference_answer": ["reference_answer/Q020"]}
{"id": 21, "task": "Given genotype file {Q021/maize_598.vcf.gz}and phenotype file {Q021/phe.csv}, first perform population structure inference. Then, conduct QEI detection using the genotype and phenotype files, setting the parameters with svrad=20000 bp, svpal=0.01, and svmlod=3. Extract significant loci based on P-values from the results and save them to loci.txt. Subsequently, extract these loci from the genotype file to generate a BED format file loci.bed. Finally, perform structural annotation of these significant QEI loci against the B73 V4 reference genome.", "reference_steps": 6, "categories": ["Population genetic structure analysis", "Environmental & Phenotypic data analysis", "Gene mining & Functional characterization"], "data_files": ["files/Q021/maize_598.vcf.gz", "files/Q021/phe.csv"], "reference_answer": ["reference_answer/Q021"]}
{"id": 22, "task": "Given genotype file {Q022/maize_598.vcf.gz}and phenotype file {Q022/phe.csv}, perform GWAS analysis using three methods, specifically FarmCPU, MLM, and GLM, to identify significantly associated genetic markers. For the FarmCPU results, extract candidate genes by defining flanking regions of 50 kb upstream and downstream of the significant markers and identifying genes residing within these intervals. During GWAS execution, specify the parameters with vc.method=“BRENT”, method.bin=static”, and threshold=0.05.", "reference_steps": 5, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q022/maize_598.vcf.gz", "files/Q022/phe.csv"], "reference_answer": ["reference_answer/Q022"]}
{"id": 23, "task": "Given genotype file {Q023/maize_598.vcf.gz}and phenotype file {Q023/phe.csv}, calculate the phenotypic correlation across different regions and visualize it using a heatmap. Concurrently, calculate the GRM based on the genotype file.", "reference_steps": 3, "categories": ["Environmental & Phenotypic data analysis", "Population genetic structure analysis"], "data_files": ["files/Q023/maize_598.vcf.gz", "files/Q023/phe.csv"], "reference_answer": ["reference_answer/Q023"]}
{"id": 24, "task": "Estimate the heritability of the trait using the PLINK binary genotype files (prefix “700w”) and the phenotype file {Q024/test.phe.txt}. First, construct the GRM from the genotype data, generating output files with the prefix “grm”. Then, estimate the narrow-sense heritability (h2) by fitting a mixed linear model that incorporates both the calculated GRM and the phenotypic data.", "reference_steps": 3, "categories": ["Genetic parameter estimation & Genomic prediction", "Population genetic structure analysis"], "data_files": ["files/Q024/test.phe.txt"], "reference_answer": ["reference_answer/Q024"]}
{"id": 25, "task": "Perform data quality control on {Q025/maize_598.vcf.gz}by filtering out variants with MAF < 0.01. Calculate the allele frequencies for the filtered dataset and save the results to a file.", "reference_steps": 3, "categories": ["Data quality control"], "data_files": ["files/Q025/maize_598.vcf.gz"], "reference_answer": ["reference_answer/Q025"]}
{"id": 26, "task": "Generate a missingness report for the genotype file {Q026/maize_598.vcf.gz}. Then perform quality control by removing variants with a missing rate >5%, variants failing the Hardy-Weinberg Equilibrium test (p < 1e-4), and samples with a variant missing rate >5%. Save the filtered output as new.vcf.", "reference_steps": 3, "categories": ["Data quality control"], "data_files": ["files/Q026/maize_598.vcf.gz"], "reference_answer": ["reference_answer/Q026"]}
{"id": 27, "task": "Given a genotype file {Q027/maize_598.vcf.gz}and a list of SNP loci {Q027/locilist.txt} mapped to the B73 V4 reference genome, perform structural annotation for the listed SNPs. Simultaneously, extract these specific loci from the genotype file to generate new.vcf. Convert new.vcf into a numeric matrix and encode the genotypes using an additive model (0, 1, 2).", "reference_steps": 3, "categories": ["Data quality control", "Gene mining & Functional characterization"], "data_files": ["files/Q027/maize_598.vcf.gz", "files/Q027/locilist.txt"], "reference_answer": ["reference_answer/Q027"]}
{"id": 28, "task": "Using genotype file {Q028/maize_598.vcf.gz}and phenotype file {Q028/phe.csv}, perform GWAS analysis using FarmCPU, GLM, and MLM methods, specifying the parameters with vc.method=“BRENT”, method.bin=“static”, and threshold=0.05. For the significant loci identified by FarmCPU, extract the lead SNPs and calculate their allele frequencies separately for the materials listed in group A and group B of the grouping file {Q028/group.txt}.", "reference_steps": 4, "categories": ["Gene mining & Functional characterization", "Population genetic structure analysis"], "data_files": ["files/Q028/maize_598.vcf.gz", "files/Q028/phe.csv", "files/Q028/group.txt"], "reference_answer": ["reference_answer/Q028"]}
{"id": 29, "task": "Given a list of significant GWAS loci ({Q029/locilist.txt}) mapped to the B73 V4 reference genome, along with genotype file {Q029/V4700w.vcf.gz}and phenotype file {Q029/phe.csv}, first perform structural annotation to identify SNPs located within gene bodies and filter the dataset to retain only these genic variants. Subsequently, perform Gene-Gene (G×G) interaction analysis using these filtered SNPs to identify significant interacting locus pairs. Finally, map these significant locus pairs back to their corresponding genes to construct a gene-pair interaction network and save the results to a file.", "reference_steps": 4, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q029/locilist.txt", "files/Q029/V4700w.vcf.gz", "files/Q029/phe.csv"], "reference_answer": ["reference_answer/Q029"]}
{"id": 30, "task": "Given genotype file {Q030/V4700w.vcf.gz}, phenotype file {Q030/PH_phe.csv}, and regional environmental factor metadata {Q030/meta_EF_mean.csv}, quantify the interaction effects between the SNPs listed in {Q030/locilist.txt}and the environmental factors. Subsequently, using the detailed environmental time-series data {Q030/5Envs_envParas_DAP150.txt}, experimental design metadata {Q030/Env_meta_table.txt}, and formatted phenotype records {Q030/Trait_records.txt}, implement the CERIS method to calculate EIs. Based on these indices, pinpoint the most influential time windows within the specific environmental factors exhibiting significant interactions.", "reference_steps": 6, "categories": ["Environmental & Phenotypic data analysis"], "data_files": ["files/Q030/V4700w.vcf.gz", "files/Q030/PH_phe.csv", "files/Q030/meta_EF_mean.csv", "files/Q030/locilist.txt", "files/Q030/5Envs_envParas_DAP150.txt", "files/Q030/Env_meta_table.txt", "files/Q030/Trait_records.txt"], "reference_answer": ["reference_answer/Q030"]}
{"id": 31, "task": "Construct the GRM using the genotype file {Q031/maize_598.vcf.gz}.", "reference_steps": 1, "categories": ["Population genetic structure analysis"], "data_files": ["files/Q031/maize_598.vcf.gz"], "reference_answer": ["reference_answer/Q031"]}
{"id": 32, "task": "Calculate the nucleotide diversity statistics (π) based on the genotype file {Q032/maize_598.vcf.gz}.", "reference_steps": 1, "categories": ["Population genetic structure analysis"], "data_files": ["files/Q032/maize_598.vcf.gz"], "reference_answer": ["reference_answer/Q032"]}
{"id": 33, "task": "Given the phenotype file {Q033/phe.csv}, compute the pairwise correlations among the traits.", "reference_steps": 1, "categories": ["Environmental & Phenotypic data analysis"], "data_files": ["files/Q033/phe.csv"], "reference_answer": ["reference_answer/Q033"]}
{"id": 34, "task": "Using the phenotype file {Q034/phe.csv}, calculate the BLUPs for the trait across multiple environments.", "reference_steps": 1, "categories": ["Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q034/phe.csv"], "reference_answer": ["reference_answer/Q034"]}
{"id": 35, "task": "Estimate the narrow-sense heritability (h2) using the pre-calculated GRM (prefix “grm”) and the phenotype file {Q035/test.phe.txt}.", "reference_steps": 1, "categories": ["Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q035/test.phe.txt"], "reference_answer": ["reference_answer/Q035"]}
{"id": 36, "task": "Perform variance component analysis on {Q036/provide.csv} to dissect the proportions of variance explained by G, E, and G×E.", "reference_steps": 1, "categories": ["Genetic parameter estimation & Genomic prediction"], "data_files": ["files/Q036/provide.csv"], "reference_answer": ["reference_answer/Q036"]}
{"id": 37, "task": "Perform genotype imputation on the file {Q037/Chip4w.vcf.gz} to infer and fill in missing genotype calls.", "reference_steps": 1, "categories": ["Data quality control"], "data_files": ["files/Q037/Chip4w.vcf.gz"], "reference_answer": ["reference_answer/Q037"]}
{"id": 38, "task": "For the list of significant loci {Q038/locilist.txt} mapped to the B73 V4 reference genome, first generate a BED format file locallist.bed. Remove the “chr” string prefix from the chromosome identifiers within this file. Subsequently, perform structural annotation for correspond loci.", "reference_steps": 2, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q038/locilist.txt"], "reference_answer": ["reference_answer/Q038"]}
{"id": 39, "task": "Perform GO enrichment analysis and KEGG pathway analysis for the candidate genes listed in {Q039/candidate_genes.txt}.", "reference_steps": 2, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q040/candidate_genes.txt"], "reference_answer": ["reference_answer/Q040"]}
{"id": 40, "task": "Identify and annotate the protein domains for the candidates in {Q040/candidate_genes.txt}.", "reference_steps": 2, "categories": ["Gene mining & Functional characterization"], "data_files": ["files/Q039/candidate_genes.txt"], "reference_answer": ["reference_answer/Q039"]}