Posts

Installing krepp We start by installing krepp, the easiest way is through bioconda. 1conda install bioconda::krepp 2krepp --help To perform any kind of query, you need to either build an index from reference genomes or download a public krepp index from one of our servers. It is quite straightforward to download an index and get started (jump to Downloading an existing index). Alternatively, if you have your own reference mitogenomes, you can index them from scratch (see Indexing mitogenomes).

k-mers mtDNA metagenomics distance-estimation genome-skimming read-mapping

Read more ›

I really enjoyed RECOMB 2025 in Seoul. This was my first time in East Asia. I enjoyed it. The conference coincided with Yeon Deung Hoe (perhaps on purpose?) and it’s been quite an experience. My RECOMB 2025 slides are available. …and a couple of photos: Yonsei University, Seoul, South Korea. Apparently I make this same gesture during all my talks.

Read more ›

Our paper titled “A k-mer-based maximum likelihood method for estimating distances of reads to genomes enables genome-wide phylogenetic placement” has been accepted in RECOMB 2025. I’m looking forward to my presentation in Seoul, which will be my first visit to East Asia. We introduced a technique to estimate read-to-genome distances from k-mer hits. The idea is based on the search for matching k-mers up to a certain Hamming distance in a colored k-mer index, and then finding the maximum likelihood distance based on k-mer matches and corresponding distances for each hitting reference.

Read more ›

We are starting from scratch by creating a directory for the project: 1mkdir BaitsProject-ONR 2cd BaitsProject-ONR The very first step is to download all the mitochondrion genomes available on RefSeq as of 2024/09/13, we also retrieve the corresponding catalog to later map each genome to a taxonomic group. 1# 2024-09-13 2wget https://ftp.ncbi.nlm.nih.gov/refseq/release/release-catalog/RefSeq-release226.catalog.gz 3wget https://ftp.ncbi.nlm.nih.gov/refseq/release/mitochondrion/mitochondrion.1.1.genomic.fna.gz Next step is to simply decompress everything we downloaded: 1gunzip -d mitochondrion.1.1.genomic.fna.gz 2gunzip -d RefSeq-release226.catalog.gz Convert multi-line FASTA to 2-line FASTA for easier processing:

k-mers metagenomics mitochondrial-baits taxonomic-classification

Read more ›

KRANK has been published in Genome Research (RECOMB 2024 Special Issue). The tool is available at GitHub, together with links to pre-built indexes of varying sizes, utilizing WoL reference database and a RefSeq snapshot. Using these indexes, KRANK showed superior performance in taxonomic classification and abundance profiling on CAMI benchmarking datasets to many existing methods.

Read more ›

Quantifying animal behavior is the main objective of computational ethology. This task is not straightforward. From a computational perspective, it involves mapping spatiotemporal features, such as changes and postures, to behavioral categories.

computational-ethology machine-learning time-series semisupervised-learning dimensionality-reduction

Read more ›

I gave a talk on KRANK in RECOMB 2024. My RECOMB 2024 slides are available. In summary: This was my first time visiting Boston, and I enjoyed it very much. I think the organization was much better last year, in Istanbul. Poster sessions are definitely the best place to meet relevant people. Being the first speaker early in the morning was unfortunate, many people missed the entire (or a considerable portion) talk.

Read more ›

I presented KRANK in BioSys 2024 (Workshop on Emerging Computer Systems Challenges and Applications in Biomedicine), which was held in conjunction with ASPLOS 2024 (29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems). Thanks to Yatish Turakhia and his students for the organization. My slides and the recording are available. A shot taken by Kyle Smith during the presentation, Hilton La Jolla Torrey Pines, San Diego.

Read more ›

I will do my best to maintain these. You can let me know if any of the links is dead.

Read more ›

After a long review process, CONSULT-II finally appeared in Bioinformatics. I think CONSULT-II is conceptually very exciting but as a software I am not very proud of it. The implementation is very eclectic and not user-friendly at all. I was able to re-implement it for KRANK, and KRANK’s software offers the same functionality with an improved user experience and increased scalability. Another thing is that, CONSULT-II consists of many heuristics that there are alternatives for.

Read more ›

Our paper titled “Memory-bound k-mer selection for large evolutionary diverse reference libraries” has been accepted to RECOMB 2024. Hopefully, I’ll be in Cambridge (April 29 - May 2, 2024). We introduced a tool, called KRANK, to sample k-mers from a large set of reference genomes for taxonomic identification or profiling purposes.

Read more ›

We published a paper in Bioinformatics. We describe a new approach for community detection in multi-subject and dynamic networks, in particular networks constructed from scRNA-seq data. The abstract is given below, and see the illustration of what multi-subject and dynamic networks refer to. Motivation: With the wide availability of single-cell RNA-seq (scRNA-seq) technology, population-scale scRNA-seq datasets across multiple individuals and time points are emerging. While the initial investigations of these datasets tend to focus on standard analysis of clustering and differential expression, leveraging the power of scRNA-seq data at the personalized dynamic gene co-expression network level has the potential to unlock subject and/or time-specific network-level variation, which is critical for understanding phenotypic differences.

Read more ›

We just published a paper in Natural Language Engineering after a long revision process. This work was a continuation and an extension of my undergraduate graduation project.

Read more ›

I presented in RECOMB Comparative Genomics 2023, in Istanbul (see my slides). Our paper, with Elenora Rachtman and Siavash Mirarab, CONSULT-II: Taxonomic Identification Using Locality Sensitive Hashing won the best paper award. A shot taken by Can Alkan during the presentation, İTÜ Maçka Oteli, Istanbul.

Read more ›

Relying on locality-sensitive hashing, CONSULT-II k-mers from a query set and tests whether they fall within a user-specified hamming distance of k-mers in the reference dataset. Using this invaluable information, it can remove contamination, i.e., reads that do not belong to the reference set, at least predicted to do so.

metagenomics tutorial contamination-removal locality-sensitive-hashing

Read more ›

Consider a lattice, where we place pebbles with probability $p$ at each intersection. Neighboring pebbles are considered to be connected, forming clusters of size two or more. We are mainly interested in sizes of these clusters.

networks network-clusters network-communities percolation-theory

Read more ›

Deriving categorical cross entropy is quite trivial, and could help understanding the loss function better.

Read more ›