Seqinspector: position-based navigation through the ChIP-seq data landscape to identify gene expression regulators¶

Step 1: Prepare your list¶

You should prepare your input list in one of the following formats.

chr16        30254029        30255503
chr4 49268883        49272093
chr1 145968637       145969897
...

chr10:66999617-67001617
chr18:35019861-35021861
chr14:70476252-70478252
...

Egr1
Egr2
Fos
...

ENSMUST00000165033
ENSMUST00000145936
ENSMUST00000140525
...

NM_007913
NM_010118
NM_010234
...

Step 2: Choose genome assembly¶

You are allowed to choose between Homo sapiens (hg19) and Mus musculus (mm9, mm10)

Step 3: Insert your query list¶

The optimal list length is about 50 items. Seqinspector was tested with up to 1000 items lists.

Step 4: Submit and wait¶

Press the submit button. Seqinspector will automatically convert gene symbols or transcripts into genomic intervals 2000 bp long around transcription start sites. If a gene has more than one start site, all will be used. If some genomic intervals are overlapping they are merged. Seqinspector will then compute coverage from all available tracks for your query set of genomic intervals. Calculation progress will be shown on the screen.

Step 5: Inspect list of query sets¶

After pressing the submit button your query will appear on the list under unique name (“Set_1” if it is your first submission). Additionally there is allready preloaded reference set (1000 random promoters). Statistics is always computed in comparison to reference. Therefore, one of your uploaded queries might be set as a reference by pressing “set as reference” button. You can remove or rename any of your queries.

Step 6: Calculate statistics¶

After pressing the “statistics” button table with results will be presented on the bottom of the screen. The tracks are sorted by p-value of significance. The columns are:

• Track name - it is internal id of a track, that contains short name of a transcription factor
• Query - average coverage of query set
• Background - avergae coverage of reference set
• Fold diff - fold difference between query and reference
• P value - significance of difference between query and reference datasets (calculated by t-test)
• Bonferroni - Bonferroni corrected p-value
• Stack plot - heat stacked plots presenting distribution of coverages in all query sets with respective p-values
• Histogram - visualisation of average coverage (2000 bp around center of genomic interval) for all query sets
• Genes - genomic intervals in query sets, symbols for nearest genes and coverage for these intervals
• Description - description of a track

Step 7: Inspect your results¶

You can visualize your results by pressing “stack plot” or “histogram” buttons. You can also inspect which genes have the highest coverage of a particular track by pressing “show genes” button. From “show genes” dialog you can navigate to seqinspector-one tool to inspect individual gene.

Step 8: Change options¶

Change database. In seqinspector there are two databases: (1) Mus musculus and (2) Homo sapiens. You can inspect Human tracks with your murine genomic coordinates. Mouse coordinates will be translated into Human coordinates using liftover tool. Extend query range. If 1000 bp upstream and downstream from tss is too small for you, it is possible to expande query range by using this parameter. It is only possible for queries using gene symbols and transcript ids.

Step 1: Prepare your list¶

You should prepare your input in one of the following formats.

chr16   30254029        30255503

chr10:66999617-67001617

Egr1

ENSMUST00000165033

NM_007913

Step 2: Choose genome assembly¶

You are allowed to choose between Homo sapiens (hg19) and Mus musculus (mm9, mm10)

Step 3: Insert your query¶

If you will choose to insert gene symbol there is autocomplete mechanism that starts to work after first two letters. If there are more than one transcription start sites for submitted gene user can choose between them in the selector box.

Step 4: Submit and wait¶

There is autosubmission mechanism in seqinspector-one. However, in case that something will not work you can press “compute” green button. Seqinspector will automatically convert gene symbols or transcripts into genomic intervals 2000 bp long around transcription start sites. Seqinspector will then compute coverage from all available tracks for your query and will compare to genome average.

Step 5: Inspect image¶

Image shows selected tracks and ensembl gene annotation in the selected query range. Selected tracks are presented as coverage histograms (with minimum value of 2 on the Y axis). You can choose tracks for visualisation in the results table (see below).

Step 6: Inspect results¶

After pressing the “Show statistics” button table with results will be presented on the bottom of the screen. The tracks are sorted by p-value of significance. The columns are:

• Track name - it is internal id of a track, that contains short name of a transcription factor
• Query - coverage for query range
• Background - avergae coverage of precomputed reference set
• Fold diff - fold difference between query and reference
• P value - significance of difference between query and reference datasets (calculated by z-score)
• Bonferroni - Bonferroni corrected p-value
• Stack plot - heat stacked plots presenting distribution of coverages in all query sets with respective p-values
• Description - description of a track

Step 7: Change options¶

Change database. In seqinspector there are two databases: (1) Mus musculus and (2) Homo sapiens. You can inspect Human tracks with your murine genomic coordinates. Mouse coordinates will be translated into Human coordinates using liftover tool.