Our knowledge of circumstellar disks has traditionally been based on studies of dust. However, gas dominates the disk mass and its study is key to understand the star and planet formation process. Spitzer can access gas emission lines in the mid-infrared, providing new diagnostics of the physical conditions in accretion disks and outflows. We have studied the spectra of 64 pre-main-sequence stars in Taurus using Spitzer/IRS observations. We have detected H2 (17.03, 28.22 $mu$m) emission in 6 objects, [Ne II] (12.81 $mu$m) in 18 objects, and [Fe II] (17.93, 25.99 $mu$m) in 7 objects. [Ne II] detections are found primarily in Class II objects. The luminosity of the [Ne II] line, is in general higher for objects known to drive jets than for those without known jets, but the two groups are not statistically distinguishable. We have searched for correlations between the line luminosities and different parameters related to the star-disk system. The [Ne II] luminosity is correlated with X-ray luminosity for Class II objects. The [NeII] luminosity is correlated with disk mass and accretion rate when the sample is divided into high and low accretors. We also find correlations between [NeII] luminosity and mid-IR continuum luminosity and with luminosity of the [O I] (6300 AA) line, the latter being an outflow tracer. [Fe II] luminosity correlates with mass accretion rate. No correlations were found between H2 luminosity and several tested parameters. Our study reveals a general trend toward accretion-related phenomena as the origin of the gas emission lines. Shocks in jets and outflowing material are more likely to play a significant role than shocks in infalling material. The role of X-ray irradiation is less prominent but still present for [Ne II], in particular for Class II sources, the lack of correlation between [Fe II] and [Ne II] points toward different emitting mechanisms.