Biology subjects

Celik, M. H.

Publications and source records attributed to Celik, M. H..

10 recordsLinked to original sources

bioRxiv2026

Agentic systems are adept at solving well-scoped, verifiable problems in computational biology

We introduce CompBioBench, a benchmark of 100 diverse tasks for evaluating agentic systems in computational biology. Unlike mathematics and programming, which more readily admit systematic verification, biological data are inherently noisy and open to interpretation. To enable objective evaluation without reducing tasks to prescriptive checklists, we propose a new benchmark construction strategy based on synthetic/augmented data and metadata scrambling/scrubbing of real datasets to create challenging problems that have a single ground-truth answer and require multi-step reasoning, tool use, bespoke code, and interaction with real-world external resources. The benchmark spans genomics, transcriptomics, epigenomics, single-cell analysis, human genetics, and machine learning workflows. Questions are curated by domain experts to cover a broad range of skills with varying difficulty. We evaluate leading general-purpose agentic systems starting from a bare-minimum environment, requiring them to fetch data and tools as needed to solve each problem. We find strong end-to-end performance, with Codex CLI (GPT 5.4) reaching 83% accuracy, Gemini CLI (3.1 Pro) reaching 82%, Claude Code (Opus 4.6) reaching 81%, and Claude Code (Opus 4.7) reaching 78%. On the hardest questions, Claude Code (Opus 4.6) reaches 69%, Codex CLI (GPT 5.4) reaches 59%, and Gemini CLI (3.1 Pro) reaches 49%. CompBioBench provides a practical testbed for measuring the progress of agentic systems in computational biology and for guiding future benchmark design. Data and a public leaderboard are available at https://huggingface.co/collections/Genentech/compbiobench-v1.

bioinformatics↗

bioRxiv2024

Efficient Inference, Training, and Fine-tuning of Protein Language Models

Protein language models have shown great promise in predicting protein structure, function, and the effects of missense variants on protein fitness. However, their use has been limited by the substantial computational resources required. In this work, we focus on improving the computational efficiency of protein language models (PLMs), specifically the Evolutionary Scale Modeling (ESM) family, to increase the accessibility of PLMs. By implementing optimizations such as FlashAttention and Partition-Attention, a novel technique designed to handle proteins of variable length, we achieved a 16-fold speedup in inference time and reduced memory usage by 3 to 14 times for long proteins. Additionally, 4-bit quantization applied to billion-parameter models led to a 2 to 3 times reduction in memory consumption with minimal performance loss in the missense variant effect prediction task. Training efficiency was also improved, with a 6-fold reduction in runtime achieved through activation checkpointing and the DeepSpeed Zero-Offload strategy. For fine-tuning, we employed parameter-efficient methods, enabling state-of-the-art predictions of protein properties and functions by training only the model head or a small fraction of adapter weights. For instance, we achieved a Spearmans correlation coefficient of 70% in melting point prediction and an 87% area under the precision-recall curve (AU-PRC) for transcription factor prediction. Our efficient ESM (ESME) implementation significantly lowers the barrier to using these powerful models, making them accessible to academic laboratories with limited computational resources. The ESME implementation is available on PyPI (pypi.org/project/esm-efficient) and GitHub (github.com/uci-cbcl/esm-efficient).

bioinformatics↗

bioRxiv2023

Identifying dysregulated regions in amyotrophic lateral sclerosis through DNA accessibility outliers

The high heritability of ALS and similar rare diseases contrasts with their low molecular diagnosis rate post-genetic testing, pointing to potential undiscovered genetic factors. Chromatin accessibility assays quantify the activity of functional elements genome-wide, offering invaluable insights into dysregulated regions. In this research, we introduced EpiOut, a computational toolbox to identify outliers in chromatin accessibility. These outliers represent dysregulated regions where chromatin accessibility uniquely diverges from the population baseline in a single or few samples. Annotation of accessible regions with histone ChIP-seq and Hi-C indicates that outliers are concentrated in functional loci, especially among promoters interacting with active enhancers. Across different omics levels, outliers are robustly replicated, and chromatin accessibility outliers are reliable predictors of gene expression outliers and aberrant protein levels. For example, 59% of gene expression outliers can be linked to aberration in chromatin accessibility. When promoter accessibility does not align with gene expression, our results indicate that molecular aberrations are more likely to be linked to post-transcriptional regulation rather than transcriptional regulation. Our findings demonstrate that the outlier detection paradigm can uncover dysregulated regions in rare diseases. EpiOut is open-sourced and freely available at github.com/uci-cbcl/EpiOut.

bioinformatics↗

bioRxiv2023

Systematic assessment of long-read RNA-seq methods for transcript identification and quantification

AbstractThe Long-read RNA-Seq Genome Annotation Assessment Project (LRGASP) Consortium was formed to evaluate the effectiveness of long-read approaches for transcriptome analysis. The consortium generated over 427 million long-read sequences from cDNA and direct RNA datasets, encompassing human, mouse, and manatee species, using different protocols and sequencing platforms. These data were utilized by developers to address challenges in transcript isoform detection and quantification, as well as de novo transcript isoform identification. The study revealed that libraries with longer, more accurate sequences produce more accurate transcripts than those with increased read depth, whereas greater read depth improved quantification accuracy. In well-annotated genomes, tools based on reference sequences demonstrated the best performance. When aiming to detect rare and novel transcripts or when using reference-free approaches, incorporating additional orthogonal data and replicate samples are advised. This collaborative study offers a benchmark for current practices and provides direction for future method development in transcriptome analysis.

genomics↗

bioRxiv2023

The ENCODE4 long-read RNA-seq collection reveals distinct classes of transcript structure diversity

The majority of mammalian genes encode multiple transcript isoforms that result from differential promoter use, changes in exonic splicing, and alternative 3 end choice. Detecting and quantifying transcript isoforms across tissues, cell types, and species has been extremely challenging because transcripts are much longer than the short reads normally used for RNA-seq. By contrast, long-read RNA-seq (LR-RNA-seq) gives the complete structure of most transcripts. We sequenced 264 LR-RNA-seq PacBio libraries totaling over 1 billion circular consensus reads (CCS) for 81 unique human and mouse samples. We detect at least one full-length transcript from 87.7% of annotated human protein coding genes and a total of 200,000 full-length transcripts, 40% of which have novel exon junction chains. To capture and compute on the three sources of transcript structure diversity, we introduce a gene and transcript annotation framework that uses triplets representing the transcript start site, exon junction chain, and transcript end site of each transcript. Using triplets in a simplex representation demonstrates how promoter selection, splice pattern, and 3 processing are deployed across human tissues, with nearly half of multitranscript protein coding genes showing a clear bias toward one of the three diversity mechanisms. Evaluated across samples, the predominantly expressed transcript changes for 74% of protein coding genes. In evolution, the human and mouse transcriptomes are globally similar in types of transcript structure diversity, yet among individual orthologous gene pairs, more than half (57.8%) show substantial differences in mechanism of diversification in matching tissues. This initial large-scale survey of human and mouse long-read transcriptomes provides a foundation for further analyses of alternative transcript usage, and is complemented by short-read and microRNA data on the same samples and by epigenome data elsewhere in the ENCODE4 collection.

genomics↗

bioRxiv2022

Analysis of alternative polyadenylation from long-read or short-read RNA-seq with LAPA

MotivationAlternative polyadenylation (APA) is a major mechanism that increases transcriptional diversity and regulates mRNA abundance. Existing computational tools to analyze APA have low precision because these tools are designed for short-read RNA-seq, which is a suboptimal data source to study APA. Long-read RNA-seq (LR-RNA-seq) accurately detects complete transcript isoforms with poly(A)-tails, providing an ideal data source to study APA. However, current computational tools are incompatible with LR-RNA-seq. ResultsHere, we introduce LAPA, a computational toolkit to study alternative polyadenylation (APA) from diverse data sources such as LR-RNA-seq and short-read 3 sequencing (3-seq). LAPA counts and clusters reads with poly(A)-tail, then performs peak-calling to detect poly(A)-site in a data source agnostic manner. The resulting peaks are annotated based on genomics features and regulatory sequence elements such as presence of a poly(A)-signal. Finally, LAPA can perform robust statistical testing and multiple testing correction to detect differential APA. We analyzed ENCODE LR-RNA-seq data from human WTC11, mouse C2C12 myoblast, and C2C12-derived differentiated myotube cells using LAPA. Comparing LR-RNA-seq from different platforms and library preparation methods against 3-seq shows that LR-RNA-seq detects poly(A)-sites with a performance of 75% precision at 57% recall. Moreover, LAPA consistently improved TES validation by at least 25% over the baseline transcriptome annotation generated by TALON, independent of protocol or platform. Differential APA analysis detected 788 statistically significant genes with unique polyadenylation signatures between undifferentiated myoblast and differentiated myotube cells. Among these genes, 3 UTR elongation is significantly associated with higher expression, while shortening is linked with lower expression. This analysis reveals a link between cell state/identity and APA. Overall, our results show that LR-RNA-seq is a reliable data source for the study of post-transcriptional regulation by providing precise information about alternative polyadenylation. AvailabilityLAPA is publicly available at https://github.com/mortazavilab/lapa and PyPI. Contact:: ali.mortazavi@uci.edu

genomics↗

bioRxiv2022

Aberrant splicing prediction across human tissues

Aberrant splicing is a major cause of genetic disorders but its direct detection in transcriptomes is limited to clinically accessible tissues such as skin or body fluids. While DNA-based machine learning models allow prioritizing rare variants for affecting splicing, their performance on predicting tissue-specific aberrant splicing remains unassessed. Here, we generated the first aberrant splicing benchmark dataset, spanning over 8.8 million rare variants in 49 human tissues. At 20% recall, state-of-the-art DNA-based models cap at 10% precision. By mapping and quantifying tissue-specific splice site usage transcriptome-wide and modeling isoform competition, we increased precision by three-fold at the same recall. Integrating RNA-sequencing data of clinically accessible tissues brought precision to 60%. These results, replicated in two independent cohorts, substantially contribute to non-coding loss-of-function variant identification and to genetic diagnostics design and analytics.

bioinformatics↗

bioRxiv2020

MTSplice predicts effects of genetic variants on tissue-specific splicing

Tissue-specific splicing of exons plays an important role in determining tissue identity. However, computational tools predicting tissue-specific effects of variants on splicing are lacking. To address this issue, we developed MTSplice (Multi-tissue Splicing), a neural network which quantitatively predicts effects of human genetic variants on splicing of cassette exons in 56 tissues. MTSplice combines the state-of-the-art predictor MMSplice, which models constitutive regulatory sequences, with a new neural network which models tissue-specific regulatory sequences. MTSplice outperforms MMSplice on predicting effects associated with naturally occurring genetic variants in most tissues of the GTEx dataset. Furthermore, MTSplice predicts that autism-associated de novo mutations are enriched for variants affecting splicing specifically in the brain. MTSplice is provided free of use and open source at the model repository Kipoi. We foresee MTSplice to be useful for functional prediction and prioritization of variants associated with tissue-specific disorders.

genomics↗

bioRxiv2019

Detection of aberrant splicing events in RNA-seq data with FRASER

Aberrant splicing is a major cause of rare diseases, yet its prediction from genome sequence remains in most cases inconclusive. Recently, RNA sequencing has proven to be an effective complementary avenue to detect aberrant splicing. Here, we developed FRASER, an algorithm to detect aberrant splicing from RNA sequencing data. Unlike existing methods, FRASER captures not only alternative splicing but also intron retention events. This typically doubles the number of detected aberrant events and identified a pathogenic intron retention in MCOLN1. FRASER automatically controls for latent confounders, which are widespread and substantially affect sensitivity. Moreover, FRASER is based on a count distribution and multiple testing correction, reducing the number of calls by two orders of magnitude over commonly applied z score cutoffs, with a minor sensitivity loss. The application to rare disease diagnostics is demonstrated by reprioritizing a pathogenic aberrant exon truncation in TAZ from a published dataset. FRASER is easy to use and freely available.

bioinformatics↗

bioRxiv2018

Modular modeling improves the predictions of genetic variant effects on splicing

Predicting the effects of genetic variants on splicing is highly relevant for human genetics. We describe the framework MMSplice (modular modeling of splicing) with which we built the winning model of the CAGI 2018 exon skipping prediction challenge. The MMSplice modules are neural networks scoring exon, intron, and splice sites, trained on distinct large-scale genomics datasets. These modules are combined to predict effects of variants on exon skipping, alternative donor and acceptor sites, splicing efficiency, and pathogenicity, with matched or higher performance than state-of-the-art. Our models, available in the repository Kipoi, apply to variants including indels directly from VCF files.

genomics↗