Package 'selexprepR'

Build an R-native reproducibility manifest

Description

Builds a versioned selexprep_manifest_v2 object. The reader also accepts the historical Python selexprep_manifest_v1 schema, so existing analysis records remain inspectable. Hashes are computed for every supplied existing path and are keyed relative to input_root or output_root when supplied.

Usage

build_selexprep_manifest(
  library_report,
  input_paths = character(),
  output_paths = character(),
  accession = NULL,
  bioproject_id = NULL,
  runs = character(),
  parameters = character(),
  runtime_seconds_per_stage = numeric(),
  flags = character(),
  input_root = NULL,
  output_root = NULL
)

Arguments

library_report	A selexprep_library_report.
input_paths	Existing input file paths to hash.
output_paths	Existing output file paths to hash.
accession	Optional public accession.
bioproject_id	Optional BioProject accession.
runs	Run accessions.
parameters	Named character parameters used for the run.
runtime_seconds_per_stage	Named numeric stage durations.
flags	Character QC flag names.
input_root	Optional root used to make input hash keys relative.
output_root	Optional root used to make output hash keys relative.

Value

A validated selexprep_manifest.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
report <- selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))
build_selexprep_manifest(report)

---

Fetch FASTQ files from an ENA inspection

Description

Creates a deterministic, checksum-aware download plan from an existing inspection or an accession. By default no bytes are transferred; set dry_run = FALSE to download to outdir. Each completed file is checked against ENA's MD5 value when one is supplied.

Usage

fetch_selexprep_reads(
  accession_or_inspection,
  outdir,
  dry_run = TRUE,
  overwrite = FALSE,
  timeout_seconds = 30,
  require_assigned_rounds = TRUE,
  overrides = numeric()
)

Arguments

accession_or_inspection	A character accession or a selexprep_inspection returned by inspect_selexprep_accession().
outdir	Destination directory for downloaded FASTQ files.
dry_run	Whether to return the plan without downloading.
overwrite	Whether an existing destination file may be replaced.
timeout_seconds	Positive timeout for metadata and FASTQ requests.
require_assigned_rounds	Whether to refuse ambiguous or missing round assignments before creating a download plan.
overrides	Optional named integer mapping of run accessions to manually curated round numbers.

Value

A selexprep_fetch_result list containing the run-level ⁠$plan⁠, ⁠$downloaded_files⁠, and ⁠$dry_run⁠.

Examples

inspection <- inspect_selexprep_accession('PRJDB19098')
fetch_selexprep_reads(inspection, 'raw/PRJDB19098')

---

Infer SELEX round assignments from public metadata

Description

Applies a conservative, ordered metadata cascade: structured sample attributes, sample title, library name, experiment title, and design description. A run remains unsafe when no round is found or when one field contains conflicting candidates. The function records evidence rather than using enrichment signals to repair metadata.

Usage

infer_selexprep_rounds(runs, overrides = numeric())

Arguments

runs	A data frame or S4Vectors::DataFrame containing run_accession and optional metadata columns sample_title, library_name, experiment_title, design_description, and sample_attributes.
overrides	Optional named integer vector mapping run accessions to manually curated round numbers.

Value

A S4Vectors::DataFrame with one conservative round assignment per run. round_candidates is a list-column and preserves ambiguity.

Examples

runs <- data.frame(
    run_accession = c('SRR1', 'SRR2'),
    sample_title = c('Thrombin Round 1', 'Thrombin Round 2')
)
infer_selexprep_rounds(runs)

---

Inspect public sequencing metadata at ENA

Description

Queries ENA's read-run filereport endpoint without downloading sequence data. The returned selexprep_inspection retains verbatim library strategy/source metadata and one row per run with its FASTQ URLs, sizes, and MD5 checksums.

Usage

inspect_selexprep_accession(accession, timeout_seconds = 30)

Arguments

accession	Study- or run-level ENA/INSDC accession.
timeout_seconds	Positive network timeout in seconds.

Value

A selexprep_inspection list with a run-level S4Vectors::DataFrame in ⁠$runs⁠.

Examples

inspection <- inspect_selexprep_accession('PRJDB19098')
inspection$runs

---

Plot SELEX quality-control diagnostics

Description

Renders an R-native diagnostic surface corresponding to the four QC artifacts produced by the original Python workflow. Read retention and its primer-match proxy use extraction provenance from y. Sequence lengths are shown as read-weighted, per-round fractions so rounds with different depths remain comparable. The diversity view contains unique counts, Shannon entropy, and top-100 coverage.

Usage

## S3 method for class 'selexprep_qc'
plot(
  x,
  y = NULL,
  which = c("retention", "primer_match", "length", "diversity"),
  ...
)

Arguments

x	A selexprep_qc object returned by selexprep_qc().
y	Optional SummarizedExperiment used to compute x. It supplies count distributions and extraction provenance.
which	One or more diagnostic groups: 'retention', 'primer_match', 'length', or 'diversity'.
...	Named graphical parameters passed to graphics::par().

Details

When y is omitted, summaries already retained in x are still plotted. Panels that require raw counts or extraction provenance state that the data are unavailable instead of reconstructing them from incomplete summaries.

Value

x, invisibly.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
pools <- list(round_00 = rep(paste0(p5, 'ACGTACGT', p3), 500))
result <- run_selexprep(pools, low_total_reads = 0)
plot(S4Vectors::metadata(result)$qc, result, which = 'diversity')

---

Read a LibraryReport JSON file

Description

Read a LibraryReport JSON file

Usage

read_library_report(path)

Arguments

path	Path to a JSON file written by write_library_report() or by the compatible Python implementation.

Value

A validated selexprep_library_report.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
report <- selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))
path <- tempfile(fileext = '.json')
write_library_report(report, path)
read_library_report(path)

---

Read a selexprep count table

Description

Reads the stable four-column count contract (sequence, reads, rank, rpm) from TSV, CSV, RDS, or Parquet. Parquet support is optional and requires arrow; TSV is the portable default for R-native workflows.

Usage

read_selexprep_counts(path)

Arguments

path	Path to a .tsv, .csv, .rds, or .parquet count file.

Value

A validated S4Vectors::DataFrame with a Biostrings::BStringSet sequence column.

Examples

path <- tempfile(fileext = '.tsv')
write_selexprep_counts(selexprep_count(c('ACGT', 'ACGT', 'GGGG')), path)
read_selexprep_counts(path)

---

Read sequences from a FASTQ file

Description

Reads a plain or gzip-compressed FASTQ file into a Biostrings::BStringSet. BStringSet deliberately preserves both DNA and RNA alphabets, avoiding the loss of uracil that would occur when every input is coerced to DNA.

Usage

read_selexprep_fastq(path, max_reads = NULL)

Arguments

path	Path to a FASTQ or FASTQ.gz file.
max_reads	Optional positive maximum number of records to retain.

Value

A Biostrings::BStringSet of read sequences.

Examples

path <- tempfile(fileext = '.fastq')
writeLines(c('@read_1', 'ACGU', '+', 'IIII'), path)
read_selexprep_fastq(path)

---

Read FASTQ files into round-aware pipeline inputs

Description

Converts local FASTQ paths or a completed ENA fetch result into the R-native inputs used by run_selexprep(). Files belonging to the same round and mate are concatenated in input-table order. Plain and gzip-compressed FASTQ files are supported through read_selexprep_fastq().

Usage

read_selexprep_inputs(x, round = NULL, mate = NULL, max_reads_per_file = NULL)

Arguments

x	A character vector of FASTQ paths, a data frame-like object with path and round_number columns, or a non-dry-run selexprep_fetch_result returned by fetch_selexprep_reads().
round	Optional vector of non-negative round numbers. It is required for character paths unless their vector names end in round numbers.
mate	Optional vector containing R1/R2 or 1/2. When omitted, mate labels are inferred from filenames.
max_reads_per_file	Optional positive maximum number of records read from each FASTQ file.

Details

Mate labels are inferred conservatively from conventional ⁠_R1⁠, ⁠_R2⁠, ⁠_1⁠, and ⁠_2⁠ filename suffixes. Files without a mate suffix are treated as single-end only when no R2 file is present. Paired inputs must have both mates, with equal loaded read counts, in every round.

Value

A selexprep_read_inputs list with sequences_by_round, optional paired_mate_streams, read_source, and a file-level DataFrame named files. The first three elements can be passed directly to run_selexprep(), or use run_selexprep_files().

Examples

path <- tempfile(fileext = '.fastq')
writeLines(c('@read_1', 'ACGT', '+', 'IIII'), path)
inputs <- read_selexprep_inputs(path, round = 0)
inputs$sequences_by_round

---

Read a selexprep manifest JSON file

Description

Reads and validates both the portable Python selexprep_manifest_v1 format and the R-native selexprep_manifest_v2 format.

Usage

read_selexprep_manifest(path)

Arguments

path	Path to a selexprep manifest JSON file.

Value

A validated selexprep_manifest.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
report <- selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))
path <- tempfile(fileext = '.json')
write_selexprep_manifest(build_selexprep_manifest(report), path)
read_selexprep_manifest(path)

---

Reverse-complement nucleotide sequences

Description

This function reproduces the sequence-counting behavior of the Python implementation. It accepts DNA or RNA bases, keeps input case, maps U/u to A/a, and preserves ambiguous bases such as N/n. It is vectorized over sequence.

Usage

reverse_complement(sequence)

Arguments

sequence

A character vector of nucleotide sequences without missing values.

Value

A character vector of reverse-complemented sequences.

Examples

reverse_complement(c('ACGT', 'ACGU', 'ANNT'))

---

Run the in-memory SELEX preprocessing workflow

Description

Composes primer detection, extraction, sequence counting and quality control into a single Bioconductor result. The primary output is a SummarizedExperiment with one column per round and a sparse counts assay. The LibraryReport, extraction result and QC report are retained in metadata() for reproducibility and inspection.

Usage

run_selexprep(
  sequences_by_round,
  read_source = c("R1", "R2", "R1_AND_R2", "INTERLEAVED", "UNKNOWN"),
  paired_mate_streams = NULL,
  accession = NULL,
  sampling_seed = 42L,
  max_reads_per_round = NULL,
  low_total_reads = 10000L,
  primer_5p = NULL,
  primer_3p = NULL,
  extraction_mode = NULL
)

Arguments

sequences_by_round	A named list of R1 pools, as accepted by selexprep_detect(), or a selexprep_read_inputs object.
read_source	Input read layout.
paired_mate_streams	Optional named R2 pools.
accession	Optional public-study accession stored in metadata.
sampling_seed	Seed forwarded to selexprep_detect().
max_reads_per_round	Optional detection subsampling cap.
low_total_reads	QC threshold for per-round read depth.
primer_5p	Optional manually reviewed 5-prime primer override.
primer_3p	Optional manually reviewed 3-prime primer override.
extraction_mode	Optional manually reviewed extraction-mode override.

Details

Split-primer paired-end inputs return a valid zero-row experiment because full inserts cannot be counted without read merging. Their separate trimmed sides remain available in metadata(result)$extraction.

Value

A SummarizedExperiment with the LibraryReport, extraction and QC result in its metadata.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
pools <- list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500))
run_selexprep(pools, low_total_reads = 0)

---

Run selexprepR directly from FASTQ files

Description

Convenience bridge combining read_selexprep_inputs() and run_selexprep(). File-level provenance is attached to metadata(result)$input_files.

Usage

run_selexprep_files(
  x,
  round = NULL,
  mate = NULL,
  max_reads_per_file = NULL,
  ...
)

Arguments

x	A character vector of FASTQ paths, a data frame-like object with path and round_number columns, or a non-dry-run selexprep_fetch_result returned by fetch_selexprep_reads().
round	Optional vector of non-negative round numbers. It is required for character paths unless their vector names end in round numbers.
mate	Optional vector containing R1/R2 or 1/2. When omitted, mate labels are inferred from filenames.
max_reads_per_file	Optional positive maximum number of records read from each FASTQ file.
...	Named arguments forwarded to run_selexprep(). The input sequence, paired-mate, and read-source arguments are supplied by this function and cannot be overridden.

Value

A SummarizedExperiment as returned by run_selexprep().

Examples

experiment <- run_selexprep_files(
    c(round_00 = 'round_00.fastq.gz',
      round_01 = 'round_01.fastq.gz')
)

---

Query the bundled public HT-SELEX study catalog

Description

Returns the curated catalog snapshot distributed with the package. The optional text query is matched case-insensitively against the BioProject ID, study title, and target, making it useful for discovery while leaving raw metadata values intact.

Usage

selexprep_catalog(query = NULL)

Arguments

query

Optional non-empty search string.

Value

An S4Vectors::DataFrame of curated study metadata.

Examples

selexprep_catalog('FGF-9')

---

Count unique SELEX sequences

Description

Counts the observed sequences in one primer-stripped SELEX round. The result is a S4Vectors::DataFrame compatible with the stable Python output schema: sequence, reads, rank, and rpm. The sequence column is a Biostrings::BStringSet so that both DNA and RNA alphabets can be carried without lossy coercion.

Usage

selexprep_count(sequences)

Arguments

sequences

A character vector or a Biostrings::XStringSet containing the primer-stripped sequences from one round.

Details

Ties in abundance are ordered lexicographically by sequence. The original Python implementation does not define a tie-breaker; making it explicit in R provides reproducible ranks across platforms.

Value

A S4Vectors::DataFrame with sequence, reads, rank, and rpm columns, ordered by decreasing read count.

Examples

selexprep_count(c('ACGT', 'ACGT', 'GGGG'))

---

Demultiplex SELEX reads with user-supplied barcodes

Description

Assigns reads to selection rounds from an equal-length barcode at the R1 5-prime end. Barcode sequences are supplied explicitly; they are never inferred. R2 mates follow their R1 assignment and remain untrimmed.

Usage

selexprep_demultiplex(
  sequences,
  barcodes,
  paired_mates = NULL,
  max_mismatches = 1L,
  trim_barcode = TRUE
)

Arguments

sequences	R1 sequences as a character vector or Biostrings::XStringSet.
barcodes	Named integer vector mapping barcode sequences to non-negative round numbers, for example c(AAAAA = 0L, TTTTT = 1L).
paired_mates	Optional R2 sequences in the same order as sequences.
max_mismatches	Maximum Hamming distance allowed from a barcode.
trim_barcode	Whether to remove the matched barcode from assigned R1 sequences.

Details

Barcodes must have pairwise Hamming distance of at least 2 * max_mismatches + 1, preventing ambiguous assignments. RNA U and DNA T are treated as equivalent during matching.

Value

A selexprep_demultiplex object. Its first four fields follow the selexprep_read_inputs contract and can be supplied directly to run_selexprep(). Additional fields contain unassigned reads, a summary, and the validated barcode map.

Examples

reads <- c("AAAAACCCCC", "AAAATGGGGG", "TTTTTACGTA", "GGGGGAAAAA")
demultiplexed <- selexprep_demultiplex(
    reads,
    c(AAAAA = 0L, TTTTT = 1L)
)
demultiplexed$summary

---

Detect SELEX library structure and constant regions

Description

Detects 5' and 3' constant regions in the earliest supplied SELEX round, then verifies the call across rounds. The returned LibraryReport separates inferred biology, input layout, and required downstream action. When primer evidence is insufficient, it returns an explicit safe-failure report rather than guessing.

Usage

selexprep_detect(
  sequences_by_round,
  read_source = c("R1", "R2", "R1_AND_R2", "INTERLEAVED", "UNKNOWN"),
  paired_mate_streams = NULL,
  sampling_seed = 42L,
  max_reads_per_round = NULL
)

Arguments

sequences_by_round	A named list of character vectors or Biostrings::XStringSet objects. Each element is one sequencing round.
read_source	One of 'R1', 'R2', 'R1_AND_R2', 'INTERLEAVED', or 'UNKNOWN'.
paired_mate_streams	Optional named list containing the corresponding R2 sequences. Required only to detect split-primer paired-end layouts.
sampling_seed	Integer seed used when max_reads_per_round samples input reads.
max_reads_per_round	Optional positive cap on reads used per round.

Value

A selexprep_library_report.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))

---

Extract random regions from a SELEX library

Description

Uses a selexprep_library_report to remove the inferred constant regions. The implementation operates directly on R sequence objects and preserves the explicit safe-failure behavior of the original pipeline: reports marked as unable to extract stop with an error unless the caller supplies a corrected report. In split-primer paired-end mode, the two trimmed sides remain separate; they are never silently merged.

Usage

selexprep_extract(
  sequences_by_round,
  library_report,
  paired_mate_streams = NULL,
  primer_5p = NULL,
  primer_3p = NULL,
  extraction_mode = NULL
)

Arguments

sequences_by_round	A named list of character vectors or Biostrings::XStringSet objects containing R1 sequences.
library_report	A selexprep_library_report from selexprep_detect() or read_library_report().
paired_mate_streams	Optional named R2 sequence pools, required for PAIRED_END_SPLIT_PRIMERS reports.
primer_5p	Optional manually reviewed 5-prime primer override.
primer_3p	Optional manually reviewed 3-prime primer override.
extraction_mode	Optional manually reviewed extraction-mode override.

Details

Full-primer matching follows the cutadapt defaults used by the Python pipeline: adapters may occur away from the read boundary, substitutions and indels are allowed up to a 10 percent error rate, and RNA U is matched as DNA T. The returned insert retains the alphabet and case of the input read.

Value

A selexprep_extraction. Full and single-primer modes store one Biostrings::BStringSet per round; paired split mode stores an r1 and r2 BStringSet for every round.

Examples

primer_5p <- 'GGTAATACGACTCACTATAGGG'
primer_3p <- 'CCATGCATGCATGCATGCAT'
bases <- c('A', 'C', 'G', 'T')
random_regions <- vapply(0:499, function(i) {
    paste0(bases[(i * 7 + 0:15 * 13) %% 4 + 1], collapse = '')
}, character(1))
reads <- paste0(primer_5p, random_regions, primer_3p)
report <- selexprep_detect(list(round_00 = reads))
selexprep_extract(list(round_00 = reads[seq_len(3L)]), report)

---

Curated public HT-SELEX catalog

Description

A frozen, offline snapshot of 240 public HT-SELEX deposits. Eight experimental fields were extracted independently by two language models, reconciled, and retained with a per-field curation status. The flat dataset intentionally omits evidence quotations; immutable links to the canonical provenance-rich JSON and its generation materials are stored in S4Vectors::metadata(selexprep_public_catalog).

Usage

data(selexprep_public_catalog)

Format

An S4Vectors::DataFrame with 240 rows and 19 columns:

bioproject_id

ENA BioProject accession or a platform-prefixed Figshare or Zenodo deposit identifier.

source

ena, or the platform-prefixed Figshare or Zenodo deposit identifier used as the source key.

study_title

Title recorded for the public study or deposit.

study_type, study_type_curation

Curated experiment type and its reconciliation status.

target, target_curation

Selection target and its reconciliation status.

target_class, target_class_curation

Target class and its reconciliation status.

chemistry, chemistry_curation

DNA or RNA library chemistry and its reconciliation status.

n_random, n_random_curation

Reported random-region length and its reconciliation status.

n_rounds, n_rounds_curation

Reported selection-round count and its reconciliation status.

selection_format, selection_format_curation

Reported selection format and its reconciliation status.

counter_selection, counter_selection_curation

Reported counter-selection conditions and their reconciliation status.

Curation statuses are concordant, discordant, not_stated, verified, single_source:claude, or single_source:codex.

Details

Missing values mean that a field was not stated in the curated sources. A discordant status means that both extracted values are retained in the corresponding value column, separated by ' || '.

Source

The snapshot was migrated without modification from the original selexprep catalog at commit b6792637ec9bed78f131e8f63e12e155f733e9ae. See the object metadata and inst/CATALOG_PROVENANCE.md for immutable source links and terms.

---

Quality-control summary for a SELEX count experiment

Description

Calculates depth, diversity, enrichment and sequence-length summaries for each round of a SummarizedExperiment returned by run_selexprep(). It also returns actionable flags for low depth, unstable random-region length, weak primer evidence and paired-end libraries that need read merging.

Usage

selexprep_qc(
  experiment,
  library_report = NULL,
  low_total_reads = 10000L,
  rarefaction_depth = 10000L,
  rarefaction_seed = 42L,
  kmer_size = 6L,
  kmer_top_sequences = 10000L
)

Arguments

experiment	A SummarizedExperiment with an assay named 'counts' and a sequence column in rowData.
library_report	Optional selexprep_library_report. When omitted, a report stored in the experiment metadata is used when available.
low_total_reads	Minimum number of reads expected in every round.
rarefaction_depth	Maximum common depth used when comparing observed diversity across rounds.
rarefaction_seed	Integer seed for deterministic rarefaction.
kmer_size	Length of canonical k-mers used for between-round consistency diagnostics.
kmer_top_sequences	Maximum number of abundant sequences considered from each round for k-mer diagnostics.

Value

A selexprep_qc list with per_round, round_similarity, and flags DataFrames plus the applied settings.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
pools <- list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500))
result <- run_selexprep(pools, low_total_reads = 0)
selexprep_qc(result)

---

selexprepR result objects

Description

The package uses lightweight S3 lists for decision records and reports, while sequence abundance data use the standard Bioconductor SummarizedExperiment container. These lists are deliberately explicit and serializable: they can be written to the versioned JSON contracts without loss of meaning.

Details

The list elements are stable public contracts:

• selexprep_library_report stores primer_5p, primer_3p, their variant evidence (variants_5p, variants_3p), known_adapter_hits, extraction_mode, full_insert_recovered, read_source, required_action, orientation, random-region length summaries (n_length_mode, n_length_distribution, n_length_confidence), primer match and position-consistency rates, the inference fraction and seed, overall confidence, status, and an optional failure_reason.

• selexprep_extraction stores extraction_mode, full_insert_recovered, primer-stripped sequences_by_round, named input_reads and output_reads counts, and the pre-trim forward/reverse/ambiguous strand_distribution for each round.

• selexprep_qc stores a per_round metrics DataFrame, a pairwise round_similarity DataFrame, a flags DataFrame whose evidence column contains structured lists, and the applied settings.

• selexprep_manifest stores the schema and package/runtime versions, accession and run identifiers, input/output SHA-256 maps, the embedded library report and its classification, parameters, stage durations, QC flags, and sampling seed. Readers accept portable Python v1 and R-native v2 schemas.

• selexprep_inspection stores the requested accession, BioProject ID, study title, library strategy/source, and a run-level runs DataFrame containing accessions, titles, sizes, FASTQ URLs, and checksums.

• selexprep_fetch_result stores the resolved run-level download plan, downloaded_files, and the dry_run status.

• selexprep_read_inputs stores round-aware R1 pools, optional R2 pools, the inferred read_source, and a file-level provenance DataFrame.

• selexprep_demultiplex extends selexprep_read_inputs with unassigned reads, per-round assignment summaries, and the validated barcode map.

S3 lists are used for these small decision and provenance records because their named elements map losslessly to the versioned JSON contracts. The high-dimensional assay remains a standard SummarizedExperiment with a sparse Matrix assay.

Value

A description of the result classes and their stable list elements.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
report <- selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))
class(report)

---

Write a LibraryReport as deterministic JSON

Description

Write a LibraryReport as deterministic JSON

Usage

write_library_report(report, path)

Arguments

report	A selexprep_library_report.
path	Output JSON path.

Value

The SHA-256 digest of the emitted UTF-8 JSON, invisibly.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
report <- selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))
path <- tempfile(fileext = '.json')
write_library_report(report, path)

---

Write a selexprep count table

Description

Writes the stable four-column count contract. Use TSV for a dependency-free, portable exchange format; Parquet is available when the optional arrow package is installed.

Usage

write_selexprep_counts(counts, path)

Arguments

counts	A count table returned by selexprep_count().
path	Output path ending in .tsv, .csv, .rds, or .parquet.

Value

path, invisibly.

Examples

counts <- selexprep_count(c('ACGT', 'ACGT', 'GGGG'))
path <- tempfile(fileext = '.tsv')
write_selexprep_counts(counts, path)

---

Write a selexprep manifest as deterministic JSON

Description

Write a selexprep manifest as deterministic JSON

Usage

write_selexprep_manifest(manifest, path)

Arguments

manifest	A selexprep_manifest.
path	Output JSON path.

Value

The SHA-256 digest of the emitted UTF-8 JSON, invisibly.

Examples

p5 <- 'GGTAATACGACTCACTATAGGG'
p3 <- 'CCATGCATGCATGCATGCAT'
report <- selexprep_detect(list(round_00 = rep(paste0(p5, 'ACGTACGTACGTACGT', p3), 500)))
path <- tempfile(fileext = '.json')
write_selexprep_manifest(build_selexprep_manifest(report), path)

---