Basically there are two things you could do:

1) Align your EST sequences with existing Genome databases (like ENSEMBL gene) and check the annotations of the genes found.

2) Analyse the sequences themselves to see whether they contain coding regions

Chris Evelo answer would be sufficient to predict if you have a small data of EST if you have a large data it would be tedious .. Take a look of following publications hopefully you could find out wich way to go..good luck.!

http://nar.oxfordjournals.org/content/34/2/635.full

http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2052.2012.02325.x/abstract

http://nar.oxfordjournals.org/content/25/4/781.full.pdf

Zhipeng Qu and David Adelson from the University of Adelaide published an nice pipeline a few years ago for doing exactly what you want: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0042638

Hope this help.

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0042638