Yeast RBPs

• Yeast-shared

This tutorial compares RNAs UV-crosslinked to RNA-binding proteins (RBPs) Kre33 [Sharma-2017], Puf6 [Gerhardy-2021] and Nab3 [van.Nues-2017] in yeast Saccharomyces cerevisiae. The purpose is to show how similar experiments can be used as mutual controls for detecting unspecific background. Important is that the analysis is done with at least one protein known not to be directly involved in binding the same RNA molecules as the other proteins. Here, data for proteins proven to be involved in processing and assembly of pre-rRNA, Kre33 and Puf6, are set against data for the termination factor Nab3 to identify regions of the pre-rRNA that specifically associate to Kre33 or Puf6. More detailed analysis for these proteins is presented in the referred publications.

First,

• create a work directory (say Kre33Puf6/)

Then, move some files that have been shipped with the program:

• copy the contents of this folder in the source ditribution coalispr/docs/_source/tutorials/Yeast/shared/ to Kre33Puf6/.

In a terminal, change directory to the created work environment (from which all scripts and commands will be run):

• cd /<path to>/Kre33Puf6/

Dataset

The yeast data we are using here have originally been aligned with Novoalign and analyzed with scripts from the pyCRAC suite [Webb-2014]. For evaluating the data with Coalispr we need bedgraph files. We can obtain these after aligning the data to the reference genome. To do this, the raw sequencing data is used and downloaded from the Gene Expression Omnibus (GEO) database with the relevant accession numbers retrieved from the literature:

CRAC-data	Reference	GEO acc. no.	SRA table
Kre33	[Sharma-2017]	GSE87480	GSE87480
Puf6	[Gerhardy-2021]	GSE174587	GSE174587
Nab3	[van.Nues-2017]	GSE85545	GSE85545

• Open the GEO Accession Display page for each of the experiments.

• Enter the accession no. into the field GEO accession and press GO.

For Kre33 and Puf6 all files are collected:

• Access SRA Run Selector from bottom of the GEO accession display page that has been opened.

• On the SRA Run Selector webpage, click on the Accession List button in the Total row of the Select pane.

• Save the file to the work directory and, because it will be combined with other data, add a prefix.

  Save as Kre33-SRR_Acc_List.txt and Puf6-SRR_Acc_List.txt.

• For detailed info, click the Metadata button for a file describing the experiment.

  Save as Kre33-SraRunTable.txt and Puf6-SraRunTable.txt

Collate the lists:

• cat Kre33-SRR_Acc_List.txt Puf6-SRR_Acc_List.txt > SRR_Acc_List.txt

For Nab3 select the glucose tests on the SRA Run Selector webpage:

• In the Found # items pane, select SRR4024838, SRR4024839, and SRR4024840.

• In the Select pane, click on the Selected button and then the Accession List button.

  Save as Nab3-SRR_Acc_List.txt.

• For experiment details, click the Metadata button.

  Save as Nab3-SraRunTable.txt.

Collate the lists:

• cat Nab3-SRR_Acc_List.txt >> SRR_Acc_List.txt [1]

Download, extract fastq and compress the data (These steps create a directory structure (see Mouse miRNAs) in the working folder the scripts rely on).

• sh 0_0-SRAaccess.sh

  This takes awhile [2], and so does [3]:

• sh 0_1-SRAaccess.sh

• sh 0_2-gzip-fastq.sh

Reference traces

Traces for gene-expression in cells grown under the same conditions support analysis. Coalispr can include these RNA-seq mRNA signals as reference traces. We would like to compare CRAC signals to mRNA signals. Although the next step can be included within the Nab3 downloads, often such data have to be downloaded from another GEO experiment, which we simulate here.

The mRNA reads for parental strain BY4740 linked to the Nab3 dataset will be used [van.Nues-2017]:

RNA-Seq	Reference	GEO acc. no.	SRA table
BY4741	[van.Nues-2017]	GSE85545	SRR4024831

• Obtain the metadata as above, save as BYRNASEQ-SraRunTable.txt.

Get the sequencing data (this step is optional; omit if reference traces are not needed) [5]:

• prefetch SRR4024831

• sh 0_1_2-SRAaccess_ref.sh SRR4024831

We want to align the reference reads independently from the CRAC data [6] and for that change all SRR4024831 labels to refSRR4024831. First, SRR4024831-files are renamed and then the folder:

• for i in 1 2; do mv SRR4024831/SRR4024831_$i.fastq SRR4024831/refSRR4024831_$i.fastq; done; mv SRR4024831 refSRR4024831;

Compress the uncollapsed reads.

• sh 0_2-gzip-fastq.sh refSRR4024831

Alignment

In the other tutorials, sequence alignments were created with STAR; we will do that here as well. Before the data can be aligned to the yeast genome, we have to obtain reference files. The pyCRAC suite comes with reference fasta and gtf files for yeast, defining many non-coding RNAs [4]. More up-to-date versions have been used for the referenced papers and included here [7]. The GTF file should have been copied from the shared folder. Download an accompanying, sorted fasta genome:

• wget -O - https://ftp.ensembl.org/pub/release-107/fasta/saccharomyces_cerevisiae/dna/Saccharomyces_cerevisiae.R64-1-1.dna_sm.toplevel.fa.gz | gunzip > saccer-R64.fasta

Then create the indices:

• sh 1_1-star-indices.sh yeast  saccer-R64.fasta Saccharomyces_cerevisiae.R64-1-1.75_1.2.gtf

The adapters have been removed from the reads deposited at GEO; still the 3’adapter (App_PE, see ‘Oligonucleotides for cDNA libraries’) is found in forward reads in reverse reads, maybe formed as primer-dimer, and some 5’ adapters. To remove these:

• sh 2_0-flexbar-trim.sh

Make a dataset of collapsed reads:

• sh 2_1-collapse-pre-trimmed-seqs.sh

The reads can now be aligned to the reference genome [8]. Because UV-crosslinking is an inefficient event, compared to normal RNA-IP, low numbers of RNAs are isolated that are specifically bound to the protein of interest. The covalently bonded RNA, after being released from this protein by enzymatic digestion with proteinase K [McKellar-2020], can still carry a crosslinked residue that will interfere with cDNA synthesis. These factors lead to reduced yields of specific sequences, enabling amplification of unspecific background. Isolation of cDNAs for specific RNAs is further affected when the RNA-binding protein of interest is low in abundance or its RNA-binding substrate has a strong, higher-order structure. A strategy to reduce the number of background reads in the analysis of crosslinked RNAs is to apply a high stringency at the mapping stage [9].

Due to UV-crosslinking, point-deletions and point-mutations are common in CRAC and CLIP-cDNAs. Therefore, such reads will be mapped while allowing at least an 1 nt mismatch;

Alignments and subsequent files will be stored in two directories within Kre33Puf6/ that will be created by the scripts, namely STAR-analysis1-yeast_collapsed/ and STAR-analysis1-yeast_uncollapsed/. The folder names contain the EXP parameter (i.e. ‘yeast’) which is the first argument for the mapping scripts. To align the collapsed reads with one mismatch [10] :

• sh 3_1-run-starPE-collapsed-14mer.sh yeast 1

Do the same for the uncollapsed reads, run:

• sh 4_1-run-starPE-uncollapsed-14mer.sh yeast 1

Remove the genome from shared memory (if that is used) when all mapping threads (set to 4 in the script) have completely finished:

• sh 4_1_2-remove-genome-from-shared-memory.sh yeast

And create bedgraphs for both collapsed and uncollapsed data sets by [10]:

• sh 3_2-run-make-bedgraphs.sh yeast 1

• sh 4_2-run-make-bedgraphs-uncollapsed.sh yeast 1

Note that the alignments are stored in their own subdirectories and that the filenames (Aligned.out.bam, Log.final.out etc.) are the same for each experiment. Therefore, Coalispr uses the folder names as a lead for retrieving bedgraph and bam files.

Align reference

If reference sequences have been downloaded (this was optional), these can be aligned as (un)collapsed reads and converted to bedgraphs after adding a third parameter to instruct the mapping scripts to process the reference data. Star will not see the readcounts for collapsed reads; if coverage is sufficient:

• sh 2_1-collapse-pre-trimmed-seqs.sh refSRR4024831

• sh 3_1-run-starPE-collapsed-14mer.sh yeast 1 refSRR4024831

• sh 3_2-run-make-bedgraphs.sh yeast 1 refSRR4024831

For bedgraph values reflecting both coverage and numbers of mapped reads:

• sh 4_1-run-starPE-uncollapsed-14mer.sh yeast 1 refSRR4024831

• sh 4_2-run-make-bedgraphs-uncollapsed.sh yeast 1 refSRR4024831

Coalispr

Described in the ‘How-to guides’ and like in the mouse tutorial we need to set up working conditions for the program. First a working environment is prepared and then the configuration files.

Work folder

Similar to the mouse tutorial we set up a workfolder for Coalispr inside Kre33Puf6/.

• In a terminal change directory to Kre33Puf6/ and run:

• coalispr init

• Give ‘yeast’ as the EXP name for the session, in line with the alignments created above and confirm.

• Choose the current folder for setting up the Coalispr directory.

Output shows where the configuration files are:

Configuration files to edit are in:
'/<path to>/Kre33Puf6/Coalispr/config/constant_in'
The path '/<path to>/Kre33Puf6/Coalispr' will be set as 'SAVEIN' in 3_yeast.txt.

Configuration

Experiment file

As described in the How-to guides, for Coalispr a file describing the experiments, EXPFILE, has to be created. The fields (columns) required for the program are prepared in the XtraColumns.txt copied from Yeast-shared.

For an informative overview, not all columns or rows are collected and combined with XtraColumns.txt. Also, the csv format has to be changed to that of tsv. A separate script takes care of this:

• copy python script coalispr/coalispr/resources/share/cols_from_csv_to_tab.py to the work directory Kre33Puf6/ and run:

• python3 cols_from_csv_to_tab.py -f "Kre33-SraRunTable.txt,Puf6-SraRunTable.txt,Nab3-SraRunTable.txt,BYRNASEQ-SraRunTable.txt" -t2 -e XtraColumns.txt [12]

  (-f is input file, -t 2 stands for “tutorial 2”, the one for yeast RNA binding proteins; -e for “expand with”)

The resulting file, Kre33Puf6_Exp.tsv, would be:

Run                Description   Short Category Group Method  Fraction   Experiment GEO_Accession (exp) Sample Name
SRR4305543         Kre33-data-I  K33_a        S   K33  rip33       WCE   SRX2199807          GSM2332452  GSM2332452
SRR4305544        Kre33-data-II  K33_b        S   K33  rip33       WCE   SRX2199808          GSM2332453  GSM2332453
SRR14570780     HTP-tagged Puf6   P6_a        S    P6   rip6       WCE  SRX10913999          GSM5320150  GSM5320150
SRR14570781     HTP-tagged Puf6   P6_b        S    P6   rip6       WCE  SRX10914000          GSM5320151  GSM5320151
SRR4024838     Nab3 Vari-X-link   N3_a        U    N3   rip3       WCE   SRX2016899          GSM2276892  GSM2276892
SRR4024839      Nab3 Megatron 1   N3_b        U    N3   rip3       WCE   SRX2016900          GSM2276893  GSM2276893
SRR4024840      Nab3 Megatron 2   N3_c        U    N3   rip3       WCE   SRX2016901          GSM2276894  GSM2276894
refSRR4024831    BY4741-RNA-Seq     BY        R   NaN rnaseq       WCE   SRX2016892          GSM2276885  GSM2276885

Settings file

The next file to prepare is the Coalispr/config/constant_in/3_yeast.txt within Kre33Puf66/. This was copied from the 3_EXP.txt during coalispr init above, with some fields adapted to the current analysis. Edit this file:

• scite Coalispr/config/constant_in/3_yeast.txt

  (by setting Language in the menu-bar to “Python” the active fields are highlighted compared to the comments)

Fields to be altered in the template (‘#’ indicates a comment):

• EXP : “yeast” [13]

• CONFNAM : “3_yeast.txt” [13]

• EXPNAM : “Saccharomyces cerevisiae”

• BINSTEP : 20 [14]

• USEGAPS : BINSTEP [14]

• MIRNAPKBUF : 1/4

• SETBASE : “/<path to>/Kre33Puf6/”

• MUTNO : “1” [11]

• REFNAM : “refSRR4024831_”

• REFS : REFNAM + TAG + “_” + MUTNO + “mismatch-” + EXP [15]

• EXPFILNAM : “Kre33Puf6_Exp.tsv”

• TOTAL : “rip3” [16]

• RIP1 : “rip33” [17]

• RIP2 : “rip6” [18]

• EXPERIMENT : “Description”

• MUTGROUPS{

  “K33” : “Kre33”,

  “P6” : “Puf6”,

  “N3” : “Nab3”,

  }

• METHODS{

  TOTAL:”Nab3 crac”,

  RIP1:”Kre33 crac”,

  RIP2:”Puf6 crac”,

  }

• UNSPECIFICS : [ “N3”, ]

• MUTANTS : “”

• LENGTHSNAM : “Saccharomyces_cerevisiae.R64-1-1.75_chromosome_lengths.txt”

• GTFREFNAM : “Saccharomyces_cerevisiae.R64-1-1.75_1.2.gtf”

• SAVEIN : Path(“/<path to>/Kre33Puf6/Coalispr”) [13]

• LENGTHSFILE : BASEDIR / LENGTHSNAM [19] [20]

• EXPFILE : BASEDIR / EXPFILNAM

• REFDIR : BASEDIR / SRCFLDR / REFS

• GTFREF : BASEDIR / GTFREFNAM

Analysis

After preparing the configuration files, we can process the bedgraphs.

Figure 1. rRF cDNAs and reads

Kre33 crosslinks specifically

to 18S rRNA regions; Puf6

binds 25S rRNA helices [21].

Top panel: cDNAs;

(collapsed data).

Bottom panels: reads;

(uncollapsed data).

setexp

Begin with activating the new configuration:

• cd /<path to>/Kre33Puf6/

• coalispr setexp -e yeast -p2

  choose option 2 and confirm.

storedata

Now the bedgraph data can be loaded into Pandas dataframes, saved as pickle files by Coalispr: When reference bedgraphs are used, these can be stored using option -d2. First, do this for the uncollapsed reads (the default type, -t1):

• coalispr storedata -d1

• coalispr storedata -d2

For collapsed reads, getting an idea of the different cDNAs crosslinked to the proteins, do:

• coalispr storedata -d1 -t2

• coalispr storedata -d2 -t2

showgraphs

Analysis of bedgraph traces for all uncollapsed reads with

• coalispr showgraphs -c XII -w2

or checking traces representing cDNAs with

• coalispr showgraphs -c XII -w2 -t2

shows how specific rRFs for Kre33 derive from 18S regions in the rDNA and those for Puf6 from a 25S section (top panels in Fig. 1). Zooming out from the rDNA region (bottom panel in Fig. 1) shows how noisy the RNA-seq data from cross-linked samples can be (e.g. for Kre33 or Puf6 [22]). Therefore, monitoring single nt deletions (for CRAC) or mutations (for CLIP) that will be common for crosslinked uracil residues in crosslinked RNA fragments is an important measure. The pyCRAC software suite [Webb-2014] has been specifically developed for such in-depth analysis of RNA-seq data for crosslinked RNA. The referred publications for Kre33 [Sharma-2017], Puf6 [Gerhardy-2021] and Nab3 [van.Nues-2017] describe such an analysis. Here we aimed to illustrate usage of Coalispr and show how independent experiments can form mutual controls for particular observations, as mentioned in the essay “”.

---

Notes

[1]

Instead of:

• In the Select pane, click on the Selected button and then the Accession List button.

  Save as Nab3-SRR_Acc_List.txt

• Collate the lists by

  cat Nab3-SRR_Acc_List.txt >> SRR_Acc_List.txt

One can collate directly:

• for i in SRR4024838 SRR4024839 SRR4024840; do echo $i >> SRR_Acc_List; done

[2]

feedback in the terminal is like:

2022-10-05T10:17:35 prefetch.2.11.0: 1) Downloading 'SRR4305543'...
2022-10-05T10:17:35 prefetch.2.11.0:  Downloading via HTTPS...
2022-10-05T10:17:53 prefetch.2.11.0:  HTTPS download succeed
..

[3]

feedback is like:

spots read      : 2,195,879
reads read      : 4,391,758
reads written   : 4,391,758

[4]

They are in the db/ folder of the pyCRAC source package.

[5]

Although the actual sequences need not be gathered, the metadata is used for creating the Experiment file (EXPFILE). Also the renaming and compression of the sequence files can be skipped.

[6]

RNA-seq reads for mRNA are obtained by a different protocol than those for CRAC samples; the latter being less abundant and with a shorter length need more stringent settings. Compare the STAR alignment commands in each mapping script.

[7]

These pyCRAC GTFs include additional annotations for ncRNAs. General updates to the reference genome can be found at Ensembl or the Saccharomyces Genome Database (sgd) ftp sites. Depending on the source, chromosome names can differ (‘chrI’ vs ‘I’ for example); therefore use fasta and gtf files from the same release-source for everything.

[8]

The alignment scripts are set up to load the genome and then keep it in shared memory, which works fine on a single computer. When the mapping is done on a cluster of shared computers, this can cause problems (see STAR/issues/604); edit the bash scripts 3_1 and 4_1 so that the setting NoSharedMemory is activated.

[9]

In STAR soft-clipping is default and will increase mapping scores and speed. Expect fast reductions in the number of mapped reads with the stringent alignEndsType --EndToEnd setting (compared to --Extend5pOfReads12).

[10] (1,2,3)

Add the number of allowed mismatches in the alignment as the second argument for the script: sh 3_1-run-starPE-collapsed-14mer.sh yeast 1 for 1 mismatch with the yeast genome. When mapping is redone with the same number of mismatches existent alignment folders within SRCFLDR [11] will be renamed from SRR.. to _SRR...

[11] (1,2)

The string MUTNO helps to define SRCFLDR in the configuration file; yielding STAR-analysis1-yeast_, which forms the beginning of the folder names for storing mapping data. These names end in collapsed or uncollapsed and are created by running mapping scripts 3_1 or 4_1. The ‘yeast’ portion relates to the EXP setting, which is the first parameter for the scripts. The digit in the name represents the number of mismatches allowed in alignments of reads to the reference genome. The default in the scripts is zero mismatches but this can be set by adding the number as the second argument to the alignment command [10]. The MUTNO setting has to reflect this new number to work with the correct data set.

[12]

The called script assumes the reference metadata has been downloaded and is available. Otherwise the XtraColumns.txt would not fit either.

[13] (1,2,3)

Done during coalispr init step.

[14] (1,2)

With CRAC [Granneman-2009] direct binding contacts between protein and RNA are sought; best to keep peaks as narrow as possible. Most positive peaks are localized hits, so account for small single peaks rather than longer regions. Bacause the yeast genome is small, setting the resolution (which is given by BINSTEP) to 20 is feasible.

[15]

Just uncomment this line in 3_yeast.txt. The name is a combination of set constants around the relabelled data-id, refSRR4024831 (REFNAM), provided as the third parameter for the mapping scripts, placing the outcomes in the SETBASE / SRCFLDR.

[16]

For RNA-ip after crosslinking in the Nab3-crac sample instead of “total”.

[17]

For Kre33 crac sample.

[18]

For Puf6 crac sample.

[19]

No quotes (” “) around the text as this is a reference to pre-defined constants; keep spaces around /.

[20]

File is placed next to sequence data folders, within the work-directory Kre33Puf6/.

[21]

The relative increases of Puf6-signals for the Kr33-specific 18S segment (and vice versa for the Kre33 trace over the Puf6-region in 25S) is probably due to cross-contamination when these samples were prepared in parallel and processed within the same RT and sequencing reactions. For other preps (Kre33-a) these overlaps were not observed.

[22]

Termination factor Nab3 (red traces) is involved in the control of many transcripts [van.Nues-2017] in line with the red peak signals. Kre33 binds to leucine and serine tRNAs as well as snoRNAs snR4 and snR45 [Sharma-2017], while Puf6 predominantly interacts with 25S rRNA sequences [Gerhardy-2021] and some snoRNAs in the target domain (RvN; unpublished). The number of these potential targets is far exceeded by the amount of hits indicated by the blue traces.