filter_by_count failed about cdna_cupcake

[Magdoll]

Hi @yuqiu77 ,

Please add the --dun_use_group_count option and this error should go away.

THanks,
--Liz

[juheon]

Could I use 'filter_by_count.py' for post-collapsing reads of FLNC reads (ISO-seq3)?

According to the Wiki, I need to an output of 'get_abundance_post_collapse.py' to run 'filter_by_count.py'. However, since I use FLNC reads instead of clustered reads, I can't run 'get_abundance_post_collapse.py' due to the absence of 'cluster_report.csv', which is the output of the clustering step.

I want to filter out singleton FLNC reads as ISO-seq3 clustering does

[Magdoll]

Hi @juheon ,

You can make a "mock" abundance.txt file for the FLNC collapsed result using the Linux commands below.

Suppose your collapsed output prefix is out.collapsed.xxxx. You need to generate the file output.collapsed.abundance.txt.

echo -e "pbid\tcount_fl" > output.collapsed.abundance.txt
paste <(awk '{print $1}' output.collapsed.group.txt) <(awk -F , '{print NF}' output.collapsed.group.txt ) >> output.collapsed.abundance.txt

This will generate an output that has the format:

pbid    count_fl
PB.1.1  1
PB.1.2  2
PB.1.3  2
PB.2.1  26
PB.2.2  1
PB.2.3  37
PB.2.4  2

which then allows you to run:

filter_by_count.py output.collapsed --dun_use_group_count