Dear all,

I'm trying to analyze some ChIPseq data using DiffBind. Samples has been processed in two different times

```
SampleID Name Factor IP Condition Replicates
5 PBS_Pol2 S3 Batch1 Pol2 PBS 1
6 PBS_Pol2 S9 Batch2 Pol2 PBS 2
7 C26_Pol2 S4 Batch1 Pol2 C26 1
8 C26_Pol2 S10 Batch2 Pol2 C26 2
```

I'm using the following code

```
dbObjprova <- dba(sampleSheet=samplesdbprova)
tamoxifen <- dba.count(dbObjprova)
tamoxifen <- dba.normalize(tamoxifen)
tamoxifen <- dba.contrast(tamoxifen,design="~Factor + Condition")
tamoxifen <- dba.analyze(tamoxifen)
```

however dba.analyze give me the following error:

```
tamoxifen <- dba.analyze(tamoxifen)
Applying Blacklist/Greylists...
Genome detected: Mmusculus.UCSC.mm10
Applying blacklist...
Removed: 142 of 3874 intervals.
Removed 142 (of 3874) consensus peaks.
Normalize DESeq2 with defaults...
Forming default model design and contrast(s)...
Analyzing...
Adding contrasts(s)...
Analyze error: Error in pv.DBA(DBA, method, bTagwise = bTagwise, minMembers = 3, bParallel = bParallel): Unable to perform analysis: no contrasts specified.
Unable to complete analysis.
Warning messages:
1: No contrasts added. There must be at least two sample groups with at least three replicates.
2: No contrasts added. There must be at least two sample groups with at least three replicates.
```

How I can solve this error? Thanks in advance

Giuseppe

Hi ATpoint,

Thanks for your quick answer.

So, If I understand correctly, in this condition and using DiffBind, I can't take into account the batch effect during differential analysis.

I have also try to analyze this data using EdgeR, but now I'm wondering if this is correct (sorry but I can't understand the math behind and for this reason I'm always wondering if I'm doing the right things).

Can I ask your opinion about this second approach?

After retrieve the count in each regions i have analyzed using the following code

In any case thanks a lot for your help

Giuseppe