You can use MACS for IP alone without a control. The basic idea is that the peaks will represent local areas of enrichment relative to a background. Without a control to measure the true background, the background might be considered a large nearby area. For each chromosome you can divide it into bins, and since you know how many reads map to that chromosome you can determine the number of reads to expect per bin. MACS then estimates for any given location, how many reads occur per bin when considering smaller local areas of 1k, 5k, and 10k. If you find a set of bins with more reads than are expected by chance, those define a peak. The catch is that, even in background samples, areas of "open chromatin" will look enriched (e.g. transcription start sites). There are many other peak finders that can use IP samples with no control, you can find them by looking up ChIP Seq on pubmed.

R provides a wealth of libraries and resources for manipulating this kind of data (look up bioconductor).

This might be helpful to you.

1. ChIP-Seq Data Analysis
2. A computational pipeline for comparative ChIP-seq analyses

It depends whether you have TF or histone data. For TFs you could try Cisgenome or Swembl, and for histone data you can try Scripture. Swembl and Scripture need a bit of parameter fiddling though.