Sujeevan Ratnasingham

Climate structures global associations between human footprint and arthropod diversity

The paper introduces Hubbell regression to estimate biodiversity descriptors and finds actual evapotranspiration is the largest predictor of arthropod diversity, with human activity reducing insect richness in tropical and dry zones.

Common to rare transfer learning (CORAL) enables inference and prediction for a quarter million rare Malagasy arthropods

The paper introduces CORAL, a transfer learning method that improves prediction and inference for rare species using DNA metabarcoding data from Madagascar.

Towards a Taxonomy Machine. A Training Set of 5.6 Million Arthropod Images

This paper releases a dataset of 5.6 million arthropod images for training machine learning models in taxonomic identification.

debar, a sequence-by-sequence denoiser for COI-5P DNA barcode data

debar is an R package that denoises COI-5P DNA barcode data using a profile hidden Markov model, reducing indel errors by 75-94% with a low false correction rate.

coil: an R package for cytochrome C oxidase I (COI) DNA barcode data cleaning, translation, and error evaluation.

The paper introduces the coil R package for cleaning, translating, and evaluating errors in COI-5P DNA barcode data, showing high accuracy in identifying technical errors but limitations in detecting biological contaminants.

Massive Multiplexing Can Deliver a $1 Test for COVID-19

This paper describes a low-cost diagnostic test for COVID-19 using sequence variation in the envelope protein gene, enabling analysis of 10,000 samples per run with RNA extraction, RT-PCR, and high-throughput sequencing.

Forensics and DNA Barcodes - Do Identification Errors Arise in the Lab or in the Sequence Libraries?

This study investigates whether misidentifications in DNA barcoding of insect species stem from reference libraries or query sequences, finding that errors originated from the latter and proposing a workflow to reduce them.

A Sequel to Sanger: Amplicon Sequencing That Scales

This paper evaluates SMRT sequencing on the SEQUEL platform for amplicon sequencing, finding it provides more complete coverage and reduces costs compared to Sanger sequencing.