10XGenomics / scHLAcount

Count HLA alleles in single-cell RNA-seq data
MIT License
59 stars 12 forks source link

Crash report #11

Closed rrydbirk closed 4 years ago

rrydbirk commented 4 years ago

I've successfully run 31 samples, but the last one crashes. Can you tell me what's wrong? All samples have been treated the same way.

name = 'sc_hla_count' operating_system = 'unix:Ubuntu' crate_version = '0.1.0' explanation = ''' Cause: assertion failed: self.len() - pos >= K::k(). Panic occurred in file '/home/travis/.c$ ''' method = 'Panic' backtrace = ''' stack backtrace: 0: backtrace::backtrace::libunwind::trace::haf40a01432d42c61 (0x562018522ce7) at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/backtrace-0.3.3$ backtrace::backtrace::trace_unsynchronized::h86bb21cca35f9927 at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/backtrace-0.3.3$ backtrace::backtrace::trace::h53026d916847a4e2 at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/backtrace-0.3.3$ 1: backtrace::capture::Backtrace::create::h15ee241c08d3314e (0x562018522243) at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/backtrace-0.3.3$ backtrace::capture::Backtrace::new::hfa537f24b00fbd81 at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/backtrace-0.3.3$ 2: human_panic::report::Report::new::hdcc2fe1769a7e55b (0x5620185111d7) at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/human-panic-1.0$ 3: human_panic::handle_dump::h319213eee1e267f2 (0x5620185108c7) at /home/travis/.cargo/registry/src/github.com-1ecc6299db9ec823/human-panic-1.0$ 4: sc_hla_count::main::{{closure}}::h93cf456556fa5fbd (0x562018459179) at /home/travis/build/10XGenomics/scHLAcount/<::human_panic::setup_panic macros$ 5: std::panicking::rust_panic_with_hook::hffcefc09751839d1 (0x562018648ec9) at src/libstd/panicking.rs:481 6: std::panicking::begin_panic::h51328ca360700b9a (0x5620184a3615) at /rustc/eae3437dfe991621e8afdc82734f4a172d7ddf9b/src/libstd/panicking.rs:411 7: ::get_kmer::h1b9ad17a9842b113 (0x56$ at /home/travis/build/10XGenomics/scHLAcount/<::std::macros::panic macros>:4 8: debruijn_mapping::pseudoaligner::Pseudoaligner::map_read::{{closure}}::h155fc7e3bad$ at /home/travis/.cargo/git/checkouts/rust-pseudoaligner-435331298a1cadf6/ae4aad$ 9: debruijn_mapping::pseudoaligner::Pseudoaligner::map_read::hd33cd425a0e12f0a (0x5620$ at /home/travis/.cargo/git/checkouts/rust-pseudoaligner-435331298a1cadf6/ae4aad$ 10: sc_hla_count::mapping::mapping_wrapper::h8e687a93e926ae64 (0x562018425bd2) at src/mapping.rs:336 11: sc_hla_count::_main::h623a028a6a6d278a (0x56201845ee13) at src/main.rs:198 12: sc_hla_count::main::h9607481a7611a730 (0x562018458fd3) at src/main.rs:139 13: std::rt::lang_start::{{closure}}::h194d0e8b51bf9c2b (0x56201844fb03) at /rustc/eae3437dfe991621e8afdc82734f4a172d7ddf9b/src/libstd/rt.rs:64 14: std::rt::lang_start_internal::{{closure}}::h447d8812e3ee306d (0x5620186487e3) at src/libstd/rt.rs:49 std::panicking::try::do_call::h4a61cb372364c745 at src/libstd/panicking.rs:296 15: __rust_maybe_catch_panic (0x56201864fd6a) at src/libpanic_unwind/lib.rs:82 16: std::panicking::try::hdf71f938885bca42 (0x5620186493ad) at src/libstd/panicking.rs:275 std::panic::catch_unwind::h7e85dbf162b1611a at src/libstd/panic.rs:394 std::rt::lang_start_internal::h1e06cc26b9fc25ea at src/libstd/rt.rs:48 17: main (0x5620184607d8) 18: __libc_start_main (0x7f694b32ab97) 19: _start (0x56201840c7b9) 20: (0x0)'''

cdarby commented 4 years ago

From the traceback, the error is occurring in the external code we call for read pseudoalignment. It fails to extract a substring of a read (kmer). I can't tell exactly what the issue is without knowing what read in your dataset caused the error. But based on the error, I have one guess of what could be going wrong. Among the reads scHLAcount uses for input (reads aligned to the HLA region), are there any reads that are very short (less than 20 bases, which is the kmer length used in the pseudoalignment algorithm)?

rrydbirk commented 4 years ago

Hi @cdarby, thanks for getting back to me. It's very likely that a short read is causing the error, I really just wanted to make sure I didn't do something wrong. Have a nice day