On an almost weekly basis, new research floods in documenting previously unidentified functionality of many different classes of non-coding DNA elements. This consistent trend renders untenable the dogmatic claim that our DNA is mostly junk.
A paper was recently published in the journal Nucleic Acids Research that identified a previously unknown function of non-coding (nc)RNA molecules (Isakov et al., 2012). Using a program called RandA (which stands for ncRNA Read-and-Analyze), a tool developed for the purpose of conducting deep-sequencing data analysis including profiling and differential expression analysis, the researchers investigated the expression patterns of the ncRNA transcriptome.
The authors begin their report by listing several other documented case of ncRNA functionality:

Non-coding RNAs (ncRNA) account for a large portion of the transcribed genomic output. They are a diverse family of untranslated transcripts that have crucial roles in cellular function. It has been shown, for example, that ncRNAs modulate the biogenesis and activity of ribosomes [small nucleolar RNA, (snoRNA)], repress gene expression [via microRNAs (miRNA)], facilitate mRNA splicing and regulate transcription factors [small nuclear RNA (snRNA)], alter cellular proliferation and apoptosis (small interfering RNA) and play a role in infrastructural functions (tRNA and rRNA). Not surprisingly, ncRNA have been implicated in human health and disease.

The researchers based their study on two human T cell cultures. One of these was infected by both human immunodeficiency virus (HIV) and Mycoplasma. The other one was not infected. The paper’s authors reported that Mycoplasma-derived ncRNA transcripts manifested a heightened level of transcript expression in the infected sample relative to the uninfected sample in which such transcripts were not detected. The figure below shows the “distribution of human RNA transcripts in the uninfected (A) and infected (B) samples.”

As you can see, there is a heightened level of expression with respect to tRNA, rRNA, and those species of ncRNA classified as “other.” snoRNA and miRNA, on the other hand, decreased in response to infection. In the case of miRNA, the paper explains that this “can be attributed to the suspected Dicer-suppressive effect exerted by HIV-1 Tat protein and/or TAR RNA.” The authors also report,

We further examined the most significantly decreased miRNAs [using DIANA-mirPath] and observed a noteworthy enrichment(P<0.001) of miRNA-targeted genes in the mitogen activated protein kinase (MAPK) pathway. The MAPK pathway modulates and induces HIV infectivity. Thus, we speculate that this decrease in MAPK pathway genes-targeting miRNAs could serve as a viral mechanism to induce pathway activity and subsequent increased infectivity. This requires further experimental validation.

The results of the study indicate that non-coding RNA plays an important role in sensing infection by pathogens.
Junk DNA was once a major argument against intelligent design. But, as seen here and in so much other research, it is being forced to retreat into the gaps in our ever-expanding scientific knowledge. And those gaps are steadily shrinking.