RSEM-RUN-EBSEQ(1)

💖 RSEM-RUN-EBSEQ(1) 💖

okay bestie, so this is rsem-run-ebseq, and it lives in the User Contributed Perl Documentation corner of your man pages. It's a little wrapper script, it does differential expression analysis, and honestly? It ate. Let's get into it 🫶

---

✨ NAME ✨

rsem-run-ebseq — a wrapper for EBSeq that does differential expression analysis. That's the whole vibe. No cap.

---

✨ SYNOPSIS ✨

rsem-run-ebseq [options] data_matrix_file conditions output_file

Three things go in, a whole lot of statistics come out. Periodt. 💅

---

✨ ARGUMENTS ✨

• data_matrix_file — this file is a m by n matrix. m is the number of genes/transcripts and n is the total number of samples. Every element in the matrix is the expected count for one particular gene/transcript in one particular sample. You can use rsem-generate-data-matrix to build this file out of your expression result files — she does the boring part for you 🎀
• conditions — a comma-separated list of values, where each value is how many replicates that condition has. So "3,3" means the data set has 2 conditions and each condition has 3 replicates. And "2,3,3" means 3 conditions, with 2, 3 and 3 replicates for each condition respectively. It's giving "count your reps before you commit", fr fr.
• output_file — the output file name. You pick it, girlie.

---

✨ OPTIONS ✨

• --ngvector <file> — this one hands EBSeq the grouping info it needs for isoform-level differential expression analysis. The file can be made by rsem-generate-ngvector. Turning this on is highly recommended if you're doing isoform-level differential expression analysis. (Default: off) 👑
• -h/--help — shows you the help info. The classic. The icon.

---

✨ DESCRIPTION ✨

So this program is a wrapper over EBSeq, right? It performs differential expression analysis, and it can handle two or more conditions — she's flexible. All genes/transcripts and their associated statistics get reported into one single output file. Main character energy incoming.

⚠️ But listen: this program does not control false discovery rate and it does not call differential expressed genes/transcripts. It just gives you the stats. If you want the FDR controlled, please use rsem-control-fdr after this program is finished, bestie. Don't skip that step 🙈

---

✨ OUTPUT ✨

• output_file — this file reports the calculated statistics for all genes/transcripts. It's written as a matrix, with row and column names. The row names are the genes'/transcripts' names, and the column names are for the reported statistics. Cute and labeled, as she should be ✨

Now here's the tea. If there are only 2 different conditions among the samples, four statistics (columns) get reported for each gene/transcript: "PPEE", "PPDE", "PostFC" and "RealFC".

• "PPEE" is the posterior probability (estimated by EBSeq) that a gene/transcript is equally expressed.
• "PPDE" is the posterior probability that a gene/transcript is differentially expressed.
• "PostFC" is the posterior fold change (condition 1 over condition 2) for a gene/transcript. It's defined as the ratio between the posterior mean expression estimates of the gene/transcript for each condition.
• "RealFC" is the real fold change (condition 1 over condition 2) for a gene/transcript. It's the ratio of the normalized within condition 1 mean count over the normalized within condition 2 mean count for the gene/transcript.

Fold changes are calculated using EBSeq's 'PostFC' function, and the genes/transcripts come out in descending order of their "PPDE" values. Sort queen 💅

But if there are more than 2 different conditions among the samples, the output format is different. For differential expression analysis with more than 2 conditions, EBSeq enumerates all possible expression patterns (like, which conditions are equally expressed and which ones aren't). Suppose there are k different patterns — then the first k columns of the output file give the posterior probability that each expression pattern is true. Those patterns are defined in a separate file, 'output_file.pattern'. Then column k+1 gives the maximum a posteriori (MAP) expression pattern for each gene/transcript, and column k+2 gives the posterior probability that not all conditions are equally expressed (column name "PPDE"). The genes/transcripts are reported in descending order of their "PPDE" column values. For the deets on how EBSeq works for more than 2 conditions, please refer to EBSeq's manual, she's got the lore 📖

• output_file.normalized_data_matrix — this file holds the median normalized version of the input data matrix. Glow-up complete 🌸
• output_file.pattern — this file is only generated when there are more than 2 conditions. It defines all possible expression patterns over the conditions using a matrix with names. Each row of the matrix is a different expression pattern, and each column gives the expression status of a different condition. Two conditions are equally expressed if and only if their statuses are the same. Simple as 🦋
• output_file.condmeans — also only generated when there are more than 2 conditions. It gives the normalized mean count value for each gene/transcript at each condition, formatted as a matrix with names. Each row is a gene/transcript and each column is a condition. The order of genes/transcripts is the same as 'output_file'. You can use this file to calculate fold changes between conditions you're interested in — very handy, very iykyk 💜

---

✨ EXAMPLES ✨

1. Say we're interested in isoform-level differential expression analysis and there are two conditions, each with 5 replicates. We've already collected the data matrix as 'IsoMat.txt' and generated the ngvector as 'ngvector.ngvec':

rsem-run-ebseq --ngvector ngvector.ngvec IsoMat.txt 5,5 IsoMat.results

The results will be in 'IsoMat.results', and 'IsoMat.results.normalized_data_matrix' contains the normalized data matrix. Slay 🔥

2. Say we're interested in gene-level analysis and there are 3 conditions. The first condition has 3 replicates and the other two have 4 replicates each. The data matrix is named 'GeneMat.txt':

rsem-run-ebseq GeneMat.txt 3,4,4 GeneMat.results

Four files — 'GeneMat.results', 'GeneMat.results.normalized_data_matrix', 'GeneMat.results.pattern', and 'GeneMat.results.condmeans' — will be generated. Four! Files! Obsessed 😭💖

---

and that's the whole page bestie, no crumbs left. this one's perl v5.38.2, dated 2024-04-14 — go forth and run your differential expression analysis, you've got this 👑🌈⭐