Marking bad channels
We strongly suggest you to inspect your data carefully before stepping into subsequent analisys.
In particular identifying bad channels (dead, too noisy or channels strongly affected by SQUID jumps) is the first very important step in your analisys, especially if you want to use Maxwell Filtering on your data: indeed MaxFilter give weird results if you have still badchannels in your data; moreover it does not work if you have more than 12 badchannels. To help you in this we provide some graphical user interface that run on our meg-servers. You can find a short description of it in these slide ( Extract Marking Bad Channels).
Finding a good judgement tradeoff in identifing bad-channels, especially for suspicious ones, is mainly a matter of practice. If you need help on this you can ask the MEG core team for some training.