non-overlapping options please help!

[matt_ms1]

Hello all,

Struggling with a 16s metagenomics research project using human fecal data. Have experience with QIIME and usearch, so not quite the first rodeo... couldn't get paired ends to align before realizing that the amplicon length was ~400, and sequencing provided 150bp read lengths. Oops!

Every paper I've read // project I've worked on has used paired-end reconstruction > open/close reference OTU selection. Faced with this new situation I'm at a loss for what tools to use.

Having a hard time what to do next. Does anyone have recommendations for what tools to look into? Eager to produce something for the PI, but hesitant to assume ~250bp of the amplicon length by just 'going for it' and filtering by PHRED > Closed reference OTU picking in QIIME... Any thoughts/advice/guidance would be greatly appreciated.

[GenoMax]

Have you tried BBMerge to see if there are some reads that can merge (have insert sizes smaller than expected).

You could try bbmerge.sh's tadpole "extend "modes to see if you are able to assemble some of this data.

[NGSMicro]

Have you tried QIIME2?

It provides a few higher resolution methods of clustering meaning you may be able to 'get away with' using only the forward reads.