Sample Data with Read Count Files

In this package you will find 4 readcount files and a sgRNA library FASTA file.

Sample Data with NGS FASTQ.gz Files

In this package you will find 4 .fastq.gz NGS files and a sgRNA library FASTA file.

Whole Genome Essential Genes Screen (90000 sgRNAs, TKOv1 library)

This screening data was published in

Steinhart,Z. et al. (2016)
Genome-wide CRISPR screens reveal a Wnt-FZD5 signaling circuit as a druggable vulnerability of RNF43-mutant pancreatic tumors.
Nat. Med.
                                                                                              

Sample Data with Read Count Files

In this package you will find 3 read count files and a sgRNA library fasta file.

sgRNA Library File in FASTA format

The sgRNA library file must be provided in a FASTA format. This library file is not only used to identify the genes to which your sgRNAs belong but also allows CRISPRAnalyzeR to annotate your screen correctly.

Example files for the most common sgRNA libraries can be downloaded from below at the end of the help section.

>ENSG00000006042_0_4299.561
GGAGCCCTCTGAGTTAGAAC

Sequencing Data (NGS .FASTQ files)

Your screening data can be uploaded either as gzipped NGS FASTQ (.fastq.gz) or tab-separated Readcount(.txt) files. In case you upload FASTQ files, CRISPRAnalyzeR uses bowtie2 to map the FASTQ reads against the provided sgRNA library file. More advanced users can adjust bowtie2 parameters in the FASTQ Options Box at Step 3. Since FASTQ files are large, we would ask you only upload gzip-compressed files as they are usually provided by your sequencing machine. In this case, the files should end with .fastq.gz. Otherwise, please compress them using gzip first.

Tab-Separated Read Count Files

You can upload tab-separated read count files, in which every line contains a unique sgRNA identifier as well as the number of reads assigned to it. Please note that the sgRNA identifiers must be the same as in the sgRNA library file you provide. Since CRISPRAnalyzeR performes the the normalization, you can upload raw read count files.

Example of a tab-separated read count file:

sgRNA Count
AAK1_104_0 0
AAK1_105_0 597
AAK1_106_0 145
AAK1_107_0 0
AAK1_108_0 0
AAK1_109_0 142

Extraction of your uploaded FASTQ.gz files

In case you upload FASTQ files instead of Read count files, your FASTQ data is extracted before the identification of your single sgRNA reads. For this purpose, CRISPRAnalyzeR needs to know how your FASTQ files do look like and how it can identify the barcode of each single sgRNA. By default, this barcode is the 20 nt target sequence as it is defined in the sgRNA library file you uploaded. To extract this information from your sequencing data, CRISPRAnalyzeR requires a so-called regular expression.

For your convenience, CRISPRAnalyzeR already provides you with pre-defined settings for the most common libraries and vector systems. As an alternative for more advanced users, you can also type in your own regular expression.

sgRNA library file structure

A single item within the FASTA file consists of two lines. In this case it is the the sgRNA identifier with its target sequence. The first line is used to identify the item. It always starts with a > followed by the sgRNA identifier. The second line provides the corresponding sgRNA target sequence without the PAM sequence .

How to select the correct Regular Expressions for sgRNA libraries

Regular Expressions are used to extract the gene identifier from your sgRNA identifier. This is important as we want to provide you not only a gene-based analysis, but also additional visualizations and information about your candidate gene. To find a matching regular expression, you need to know the structure of such an expression and of course, the structure of your sgRNA identifier as it is used within your library file. Regular expressions for our pre-made libraries can be found below .

CRISPRAnalyzeR provides you with pre-defined regular expressions for commonly used vector systems. Alternatively, see below how to find the correct expression for your custom library.

NGS Sequencing Files (.fastq.gz)

You can also provide raw data NGS files directly from your sequencing machine in a FASTQ format. Since FASTQ data can be large, the files which must be compressed using GZIP (.fastq.gz file ending). CRISPRAnalyzeR extracts all information from your FASTQ NGS file and performs the mapping against the sgRNA library file. Please make sure the FASTQ extraction pattern in Step3 matches the vector system you used for screening.

Example of a FASTQ file

@M01100:63:000000000-AMJC7:1:1101:21768:1401 1:N:0:1
TTGGATTCTTGTGGAAAGGACGAAACACCGTTCCCTGCAGCCCTCATGCGTTTTAGAGCTAGAAAT
+
CCCCCGGGGGGGGGGGGGGGGGGFGGGGGGGGGGGGGGFGFDGGGGGGGGGGGGGGGGG

CRISPRAnalyzeR needs to know which plasmid you used

In order to find the sgRNA target sequence (which is used as a barcode to identify the sgRNA), CRISPRAnalyzeR needs to know where to find it within your sequencing data. For this purpose, regular expressions are used to extract this information from your sequencing rawdata.
You can have a look at the plasmid map of the vector you used and try to identify the left and right flanking sequences next to the sgRNA target sequence (as illustrated below). Alternatively you can use one of the pre-defined settings depending on your vector system.

You can find the required regular expression for your screening vector the following way

Regular Expressions are used to extract the sgRNA target sequence barcode from your NGS FASTQ file. To find a matching regular expression, you need to know the sequence or name of the vector you used for performing the lentiviral transduction. All you need to do is select a couple of bases flanking your sgRNA target sequence on the left and right side. CRISPRAnalyzeR provides with pre-defined settings for the most common screening systems (see below), but you can use your custom system as well.
In this case, select CUSTOM in Step3 and enter a custom regular expression by the addition of 2-4 bases of both the left and right flanking sequences on the left and right site of the bracket. For the above shown example you could use the following regular expression for a sgRNA target sequence of either 20 or 21 nt in length:

The expression to detect the sgRNA target sequence is shown in blue color. Don't hesitate to ask us for help. CACC (.{20,21}) GTTTT

You can modify the pre-defined settings or add your own custom regular expression

Just tick the checkbox and enter a custom regular expression.

You can use the pre-defined settings for the most common screening vectors:

How can I download read count files?

You can download all read count files CRISPRAnalyzeR created for you.

CRISPRAnalyzeR gives you the opportunity to download the read count files for your samples.

In case you uploaded FASTQ.gz data, the read count files have been generated by mapping your sequencing data against your sgRNA library. If you come back to CRISPRAnalyzeR, you can then just upload the read count files together with your sgRNA library file, which is much faster.

The read count files are tab-separated text files and come within a gzip archive.

How can I download a quality report for my FASTQ.gz sequencing data?

CRISPRAnalyzeR also provides an extensive quality report for your sequencing data.

Once you upload FASTQ.gz sequencing data, CRISPRAnalyzeR will create a sequencing data quality report. The report is generated using the Rqc package

Welliton Souza and Benilton Carvalho (2016).
Rqc: Quality Control Tool for High-Throughput Sequencing Data.
R package version 1.6.2. https://github.com/labbcb/Rqc

What information is provided in the overview table?

How can I setup my treatment groups?

CRISPRAnalyzeR will always compare two treatment groups

You can increase the number of groups, but the minimum required number is 2. If you like you can also name each group.

How can I assign my samples to the groups?

You can assign each file to one group by clicking on the sample name. If you like to have more than one file per group, keep the CMD or CTRL button pushed while selecting the samples. Please note that your are not allowed to use the same sample in more than one treatment group.

Why do I need to set gene identifiers?

CRISPRAnalyzeR needs to know the type of gene identifier used for the screen - and can automatically convert it for you

CRISPRAnalyzeR needs to know in which organism your screen was performed and which gene identifier is used in the sgRNA library file. With this information, CRISPRAnalyzeR can offer you extensive gene annotation and visualizations.

CRISPRAnalyzeR expects the type of organism, the gene identifier and which identifier you would like to convert it to.

Supported Gene Identifiers

What are positive controls?

Positive controls are genes, for which the screening outcome is known. These will tell you whether your screening setup worked and are very important to judge your screening performance.

Which gene can be used as a positive controls strongly depends on the type of screen. For Viability/Dropout screen, essential genes can be used as positive control.

What are non-targeting contols?

Non-targeting controls are articicial genes, which are not present in the screened organism. Since these controls don't have any target in the screening organism, they can be used to judge the screening variability. Common non-targeting controls are GFP or random sgRNA sequences.

Cycle-sepcific quality plots

GC content

Shows the GC-base content for each sequencing cycle

Quality Distribution

Shows the quality score proportion per cycle. Colors are presented in a gradient Red-Blue, where red identifies calls of lower quality.

Average Quality

Describes the average quality score for each cycle of sequencing

Basecall Proportion

Describes the proportion of each nucleotide present at every cycle of sequencing.

Read Frequency and Width

Read Frequency

Shows the proportion and frequency of sequencing reads.

Read Width

Describes the length of reads within the sequencing file.

Overview

Mean

The mean read count of all sgRNAs/Genes in the dataset

Median

The median read count of all sgRNAs/Genes in the dataset

Min

The minimum read count of all sgRNAs/Genes in the dataset

Max

The maximum read count of all sgRNAs/Genes in the dataset

SD

The calculated Standard Deviation of all sgRNA/Gene read count in the dataset

Missing

The number of sgRNAs/Genes which have a read count of 0. In case the uploaded sgRNA FASTA library contains more items than the screening data, these sgRNAs/Genes will reported as missing.

Readcounts

The Readcounts tab allows you to browse through your readcount data. You can also sort according to the different columns and search for specific genes or sgRNAs. On the left you can select whether you wish to browse readcounts per sgRNA or get a summed-up readcount per gene.

Distribution

Described the log2 read count distribution for all sgRNAs of each dataset.

Did you know?

You can zoom into the plot by keeping the left-mouse button pushed. Moreover, you can select/deselect any dataset by clicking on the dataset name below the plot.

Controls

If you have set either positive or non-targeting controls in the Set Analysis Parameter section, read statistics for these controls are shown.

Read Depth

Shows the read count of each gene divided by the number of sgRNAs that contributed to it. This is used to give you an idea of read count depth normalized to the number of sgRNAs. The Read Depth plot shows you how many reads fall onto every gene present in your dataset. Positive controls (if set) are highlighted in red, non-targeting control (if set) are highlighted in blue color.

Missing sgRNAs/Genes

The missing sgRNA/Genes tab shows you how many sgRNAs/Genes were not present in the sequencing data compared to the uploaded sgRNA library file. This means the read count for this particular sgRNA or the sum of sgRNA read counts for a particular genes is 0. If a read count threshold has been applied, removed sgRNAs are NOT listed as missing.

Missing Genes

Number of Genes in which all sgRNAs were not present in the dataset compared to the provided sgRNA library.

Missing sgRNAs

Number of sgRNAs which revealed a read count of 0 in the dataset.

% of missing non-targeting sgRNAs

The percentage of sgRNAs, which belong to the non-targeting control, that were not present in the dataset.

% of missing positive control sgRNAs

The percentage of sgRNAs, which belong to the positive control, that were not present in the dataset.

Designs per Gene

Distribution of sgRNA presence for all genes in the dataset. In case all sgRNAs for all genes are present, you will expect only one bar.

Overview

A fast overview

Here you will find a boxplot visualization of the log2 Foldchanges or Z-Ratios between all uploaded samples on both gene and sgRNA level. This gives you a fast overview to check for outlying samples, that might have an unexpected influence on the analysis of your data.

Pairwise Comparisons

For a more detailed view

These plots allow you to directly compare two samples by both log2-Foldchange and Z-Ratio. With this you can inspect how foldchanges differ between your screning replicates.

Scatterplot Matrices

A fast overview

The grid gives you a fast overview of the correlation of your replicates. You can view the correlation for both sgRNA and Gene read count. CRISPRAnalyzeR calculates the Pearson as well as the Spearman coefficient. For a more detailed view, have a look at the compare datasets tab.

Scatterplots

For a more detailed view

If you are interested in a more detailed view, you can select two datasets or treatment groups and compare them on either gene or sgRNA level. You can highlight positive or non-targeting controls which will shown in red or blue color respectively. Moreover you can select multiple genes which will be highlighted in orange color within the scatterplot. For gene-based levels, please tick the 'Show the plot with gene level read count'. Moreover, you can get all scatter plots with log10-transformed data.

Which hit calling algorithms are implemented in CRISPRAnalyzeR?

CRISPRAnalyzeR offers you six different hit calling algorithm implementations to give you the chance to find the most robust hit within your screen. Since each algorithm performs the hit calling in a different way, we suggest you to have a look to all of them.

Before CRISPRAnalyzer applies hit calling, all data is normalized using DESeq2.

What visualization does CRISPRAnalyzeR offer?

CRISPRAnalyzeR offers you three different visualizations for all hit calling methods that have been implemented.

• Ranked P-Values by significance
• Distribution of unadjusted P-Values
• Log2-transformed foldchanges plotted against the -log10 transformation of the corresponding adjusted p-values

How can I find overlapping hit candidates?

Select the analysis methods to see overlapping hit candidates

In case you select an analysis method using the check boxes below, CRISPRAnalyzeR includes its results to show overlapping hit candidates. Genes are only considered as overlapping candidates between the methods, if they are below the p-value threshold of each method that you have set in the analysis settings. If there are no overlapping genes, reduce the number of included analysis methods and check the individial hit calling candidates of each analysis method.

Venn Diagram

A Venn diagram provides you the information, how many genes overlap in the different analysis algorithms. The number within a shared area indicates the number of overlapping candidates between these methods. A detailed explanation on Venn diagrams can be found on Wikipedia.

Once you select a gene, CRISPRAnalyzeR will gather all information for you

Why does CRISPRAnalyzeR reannotate sgRNAs?

CRISPRAnalyzeR reannotates every sgRNA within your screen using E-CRISP. With this information, CRISPRAnalyzeR can generate additional information and plots for, e.g. the genomic model or potential genomic binding sites.

Gene Information

General information about the selected gene are retrieved from Ensembl, EnrichR and KEGG.

Gene/sgRNA Model

CRISPRAnalyzeR creates a genomic view similar to a genome browser, which includes gene or sgRNA information.

Published Screens

You can check whether you gene of interest has been used in previously published CRISPR screens. Moreover, information about observed phenotypes is presented.

Gene Peformance

CRISPRAnalyzeR gives you a brief overview of how the selected gene performed in your screen compared to all other genes.

COSMIC Mutation Database

CRISPRAnalyzeR retrieves additional information about somatic cancer mutation from the Sange COSMIC database. Please visit the COSMIC website to find more information about the COSMIC database.

Gene Ontology

CRISPRAnalyzeR retrieves additional information from the the Gene Ontology Consortium. Please visit the Gene Ontology website to find more information about Gene Ontology.

The sgRNA section of the In-Depth Analysis allows you to have a closer look the performance and properties of individual sgRNAs targeting the selected gene.

Can I use several genes?

Yes, you can select single or multiple genes to start gene set analysis.

How is the Gene Set Analysis performed?

Gene Set Analysis is carried out using the Enrichr API available at the Enrichr website.

CRISPRAnalyzeR presents you each analysis with the Enrichr Combined Score, which is a combination of both the p-value and z-score:

If you want to learn more about Enrichr, you can also check out the following publication:

Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, Clark NR, Ma'ayan A. Enrichr:
interactive and collaborative HTML5 gene list enrichment analysis tool.
BMC Bioinformatics. 2013;128(14)

Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, Koplev S, Jenkins SL, Jagodnik KM, Lachmann A, McDermott MG, Monteiro CD, Gundersen GW, Ma'ayan A.
Enrichr: a comprehensive gene set enrichment analysis web server 2016 update.
Nucleic Acids Research. 2016; gkw377.

Pathways

CRISPRAnalyzeR checks the following databases for pathway information:

WikiPathways 2016

Wiki Pathway Database.

KEGG 2016

KEGG Database.

Biocarta 2016

Biocarta Database.

Reactome 2016

Reactome Database.

NCI-Nature 2016

NCI Nature Database.

Panther 2016

Panther Database.

Transcription

Gene Set Analysis is performed using different databases for transcriptional information provided by the Enrichr service

ChEA 2015

Enrichr Download Page.

TRANSFAC and JASPAR PWMs

JASPAR Transcription Factor Binding Sites.

ENCODE and ChEA Consensus TFs from ChIP-X

Download Dataset from Enrichr.

TargetScan microRNA Targets

TargetScan Website.

Transcription Factor PPIs

Download Dataset from Enrichr.

Ontologies

CRISPRAnalyzer uses Enrichr to get Gene Ontology Gene set analysis from the Gene Ontology Consortium.

GO Biological Process 2015

Gene Ontology Website.

GO Cellular Component 2015

Gene Ontology Website.

GO Molecular Function 2015

Gene Ontology Website.

Diseases

CRISPRAnalyzeR asks Enrichr to retrieve data from the Online Mendelian Inheritance in Man® Database.

OMIM

Online Mendelian Inheritance in Man®.

Cell Types

Retrieve genetic information about cell lines.

Cancer Cell Line Encyclopedia

Cancer Cell Line Encyclopedia.

NCI-60 Cancer Cell Lines

BioGPS Portal.

Protein Interactions

You can get a protein interaction network using the String Database.

String Database

String Database.

Compare the sgRNA populations of different genes using violin plots.

Find out more about violin plots at Wikipedia.

Select multiple genes as a variety of external information to annotate your genes of interest.

Which Annotations are available?

CRISPRAnalyzeR offers you all annotations available via the Ensembl biomaRt service

just start typing what you would like to know and CRISPRAnalyzeR will show you mathcing annotations directly. Moreover you will find a list of all available annotations below.