schema_version = "1.4" # what the separate verifier container receives from the agent's container artifacts = ["/workspace/output"] [task] name = "prima-mente/snisor-coupling" version = "1.0.0" description = "From single-nucleus PacBio long reads of human frontal cortex (SnISOr-Seq; Hardwick et al., Nature Biotechnology 2022, GEO GSE178175), estimate in excitatory neurons the within-molecule coupling (log odds ratio theta) of neighbouring alternative exons, the number of pair-reporting molecules, the molecule yield of two withheld SMRT cells (graded against the realised value), and the mean and spread over 1,609 genes of a gene-level exon coupling. Semi-synthetic: real molecules, barcodes, UMIs, SMRT cells and read structures; inclusion status of the target exons re-drawn with known coupling." keywords = ["computational-biology", "long-read", "single-nucleus", "alternative-splicing", "exon-coupling", "PacBio", "deduplication", "overdispersion", "mantel-haenszel", "scientific-inference"] [[task.authors]] name = "Prima Mente" [metadata] author_name = "Prima Mente" difficulty = "hard" category = "computational_biology" tags = ["SnISOr-Seq", "single-nucleus", "long-read", "splicing", "UMI-deduplication", "beta-binomial", "intraclass-correlation", "common-odds-ratio", "gene-id-versions", "semi-synthetic"] expert_time_estimate_min = 240.0 junior_time_estimate_min = 1200.0 source_paper = "Hardwick et al., Nature Biotechnology (2022), doi:10.1038/s41587-022-01231-3" data_accessions = ["GEO GSE178175", "SRA PRJNA737508", "github.com/noush-joglekar/sn-code"] [verifier] timeout_sec = 300.0 environment_mode = "separate" [verifier.environment] cpus = 1 memory_mb = 2048 workdir = "/workspace" network_mode = "no-network" [agent] timeout_sec = 1800.0 network_mode = "no-network" [environment] docker_image = "prima-mente/snisor-coupling:1.0" cpus = 16 memory_mb = 98304 storage_mb = 20480 workdir = "/workspace" network_mode = "no-network" build_timeout_sec = 1800.0 os = "linux"