pith. sign in

arxiv: 2601.14577 · v2 · pith:7OZNK4BPnew · submitted 2026-01-21 · 🧬 q-bio.QM

FBApro: A fast, simple linear transformation for diverse metabolic modeling tasks

Ariel Bruner , Mona Singh This is my paper

Pith reviewed 2026-05-22 12:09 UTC · model grok-4.3

classification 🧬 q-bio.QM
keywords flux balance analysismetabolic modelingsteady-state subspaceorthogonal projectionlinear transformationquadratic programmingcancer cell lines
0
0 comments X

The pith

FBApro finds the closest steady-state flux vector to any reference using one linear operation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces FBApro as a way to project a reference flux vector onto the linear subspace of steady-state fluxes that satisfy mass balance in a metabolic model. It handles cases where only some fluxes are given and where certain reactions have fixed values, all without needing to maximize a biological objective like growth. Although the underlying problem is a quadratic program that minimizes distance, the authors derive a closed-form solution that reduces to a single linear transformation based on orthogonal projections. This approach is fast to compute and simple to implement for many different modeling tasks. A reader would care because it removes the need to select and optimize an objective function while still respecting the core steady-state assumption of constraint-based modeling.

Core claim

For any given vector of reference fluxes, FBApro finds the closest flux vector within the steady-state subspace, and accounts for both partially given reference fluxes and exact constraints on reactions. While FBApro is the solution to a quadratic program, it can be implemented as a single linear operation using orthogonal projections to corresponding affine spaces and sets of linear equations.

What carries the argument

Orthogonal projection of a reference flux vector onto the affine space defined by the model's steady-state linear constraints, expressed in closed form as a linear map.

Load-bearing premise

The flux distribution that is closest in distance to the reference vector inside the steady-state subspace is the one that best represents the biological state.

What would settle it

Direct comparison of FBApro-projected fluxes against measured fluxes in cancer cell line experiments where the projected values deviate systematically from independent validation data.

Figures

Figures reproduced from arXiv: 2601.14577 by Ariel Bruner, Mona Singh.

Figure 1
Figure 1. Figure 1: Amortized running time per sample, averaged over five replicas for one sample for benchmark methods, and five replicas for 100 samples for FBApro variants, for different methods on synthetic data generated from four model organism metabolic models. 3 Results 3.1 FBApro is Fast and Scalable For FBApro to serve as a new, flexible tool in metabolic modeling, it needs to run efficiently. Bulk RNA-seq datasets … view at source ↗
Figure 2
Figure 2. Figure 2: Spearman correlations between actual fluxes and predicted fluxes on synthetic datasets derived from the four model organism metabolic models. (a) Average performance across 10 samples for each model and the unmeasured 90% of the model reactions when input data consists of a subset of reactions with noisy flux measurements and the remainder of reactions with no measurements. No data is shown for FBA as it y… view at source ↗
Figure 3
Figure 3. Figure 3: Per sample correlations between model outputs and synthetic data derived from four model organism metabolic models, while varying either the fraction of reactions with measured fluxes or the amount of noise. (3a), (3b) Results for the missing/noisy experiment (where some reactions have no information about fluxes and other reactions have noisy flux measurements). (3c), (3d) Results for the noisy/exact expe… view at source ↗
Figure 4
Figure 4. Figure 4: Performance of methods when predicting fluxes from GE data and masked flux data. 4 Discussion In this work we present FBApro, a simple linear transformation solving the problem of finding a closest steady￾state flux to a reference flux vector. We demonstrate that FBApro is computationally scalable to large amounts of data, allowing a qualitative difference in the kinds of data and in vitro experiments that… view at source ↗
Figure 5
Figure 5. Figure 5: Running time of methods on synthetic data for four model organism models [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Running time of methods on different devices and batch sizes(when applicable), synthetic data on Recon1 [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Running time of FBApro variants as a function of batch sizes on GPU synthetic data on Recon1 [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Time for setup [PITH_FULL_IMAGE:figures/full_fig_p018_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Per reaction correlations between model outputs and synthetic data on Recon1, while varying the fraction of known reactions or power of noise. 9a and 9b show the missing/noisy experiment, whereas 9c and9d show the noisy/exact experiment [PITH_FULL_IMAGE:figures/full_fig_p019_9.png] view at source ↗
read the original abstract

Constraint-based metabolic modeling is the predominant framework for simulating cellular metabolism. The central assumption of these models is that metabolism operates at a steady state, meaning that the production and consumption rates of each metabolite are balanced. This assumption imposes linear constraints on the fluxes of biochemical reactions. Flux Balance Analysis (FBA), a fundamental method in the field, is formulated as an optimization problem maximizing a cellular objective (e.g., growth) over the resulting linear subspace of steady state fluxes. Many other methods in the field are expressed either as a modification to FBA, or use FBA as a black box within an algorithm. Here, we propose a general alternative to optimization called FBApro. For any given vector of reference fluxes, FBApro finds the closest flux vector within the steady-state subspace, and accounts for both partially given reference fluxes and exact constraints on reactions. While FBApro is the solution to a quadratic program, we show that it can be implemented as a single linear operation using orthogonal projections to corresponding affine spaces and sets of linear equations. The overall approach is computationally efficient, does not require a cellular objective, and is easy to implement. We formally derive the closed-form expressions for FBApro and simpler variants, and validate it on both synthetic and real cancer cell line data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript introduces FBApro as a method that, for a given reference flux vector, computes the closest flux vector in the steady-state subspace defined by the stoichiometric matrix S (i.e., solving the quadratic program min ||v - r||_2 s.t. S v = 0, with extensions for partial references and exact equality constraints on reactions). It derives closed-form expressions showing that this projection can be implemented as a single linear operation via orthogonal projections onto the nullspace of S or corresponding affine spaces, and validates the approach on synthetic data plus real cancer cell line models. The method is positioned as a fast, objective-free alternative to optimization-based techniques such as FBA for diverse metabolic modeling tasks.

Significance. If the central claim holds, FBApro would provide a computationally efficient, linear-map alternative for flux projection tasks in constraint-based modeling, enabling rapid handling of partial data or exact constraints without repeated optimizations or specification of a biomass objective. The explicit closed-form derivation and validation on both synthetic and real datasets are strengths supporting reproducibility and potential utility in large-scale models.

major comments (1)
  1. In the formal derivation (as referenced in the abstract and the section on formal derivation and validation), the orthogonal projection onto the steady-state subspace (or its affine translate for exact constraints) solves the stated QP but does not incorporate the standard inequality bounds lb ≤ v ≤ ub. Consequently, the resulting fluxes can violate irreversibility or capacity constraints, and the cancer cell line validation only confirms geometric closeness rather than biological feasibility under the full polytope.
minor comments (1)
  1. The abstract and derivation sections would benefit from an explicit statement of the scope regarding inequality constraints to better delineate when the linear map is guaranteed to produce valid fluxes.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and insightful comments on our manuscript. The major comment raises an important point about the scope of the projection, and we address it directly below. We have revised the manuscript to incorporate clarifications on this limitation.

read point-by-point responses

  1. Referee: In the formal derivation (as referenced in the abstract and the section on formal derivation and validation), the orthogonal projection onto the steady-state subspace (or its affine translate for exact constraints) solves the stated QP but does not incorporate the standard inequality bounds lb ≤ v ≤ ub. Consequently, the resulting fluxes can violate irreversibility or capacity constraints, and the cancer cell line validation only confirms geometric closeness rather than biological feasibility under the full polytope.

    Authors: We agree with the referee that the FBApro derivation solves the unconstrained (or equality-constrained) quadratic program min ||v - r||_2 s.t. Sv = 0 (or with additional exact equalities), without enforcing the inequality bounds lb ≤ v ≤ ub that define the feasible flux polytope. This is by design: the method is intended to provide a closed-form linear transformation for rapid projection onto the steady-state subspace, which can be computed via a single matrix multiplication without iterative optimization. Incorporating bounds would convert the problem into a quadratic program with inequalities, which in general lacks a simple closed-form linear solution and would require numerical QP solvers, undermining the computational advantages highlighted in the paper. The synthetic and cancer cell line validations confirm that the projected fluxes are the nearest points in the affine subspace (i.e., geometric closeness), as stated in the manuscript. For use cases requiring strict adherence to bounds, FBApro can serve as an efficient initial projection or warm start, after which bounds can be enforced via clipping, post-processing, or a subsequent optimization step. We have added a new paragraph in the Discussion section of the revised manuscript explicitly acknowledging this scope limitation, contrasting it with standard FBA, and outlining practical ways to combine FBApro with bound-aware methods. revision: yes

Circularity Check

0 steps flagged

No circularity: derivation applies standard orthogonal projection to steady-state nullspace

full rationale

The paper derives FBApro directly from the definition of the Euclidean closest point in the affine space defined by S v = 0 (plus exact equality constraints). Closed-form linear expressions follow from the standard formula for orthogonal projection onto the nullspace of S, which is a textbook linear-algebra result independent of the paper's own data or prior claims. No parameters are fitted and then renamed as predictions, no self-citation chain carries the central step, and no ansatz is smuggled in. The method is self-contained against external linear-algebra benchmarks; validation on synthetic and cancer-cell data tests application rather than the derivation itself.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard domain assumption of steady-state metabolism and basic linear-algebra facts about orthogonal projections; no free parameters or new invented entities are introduced in the abstract.

axioms (1)
  • domain assumption Metabolism operates at steady state, imposing linear constraints on reaction fluxes.
    Explicitly stated as the central assumption of constraint-based metabolic modeling in the abstract.

pith-pipeline@v0.9.0 · 5759 in / 1191 out tokens · 60199 ms · 2026-05-22T12:09:49.325026+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

42 extracted references · 42 canonical work pages

  1. [1]

    Equations and calculations of product yields and preferred pathways for butanediol and mixed-acid fermentations.Biotechnology and bioengineering, 27(1):50–66, 1985

    Eleftherios Terry Papoutsakis and Charles L Meyer. Equations and calculations of product yields and preferred pathways for butanediol and mixed-acid fermentations.Biotechnology and bioengineering, 27(1):50–66, 1985

    work page 1985

  2. [2]

    What is flux balance analysis?Nature biotechnology, 28(3):245–248, 2010

    Jeffrey D Orth, Ines Thiele, and Bernhard Ø Palsson. What is flux balance analysis?Nature biotechnology, 28(3):245–248, 2010

    work page 2010

  3. [3]

    Fat synthesis in adipose tissue

    David A Fell and J Rankin Small. Fat synthesis in adipose tissue. an examination of stoichiometric constraints.Biochemical journal, 238(3):781–786, 1986

    work page 1986

  4. [4]

    Bigg models 2020: multi-strain genome-scale models and expansion across the phylogenetic tree.Nucleic Acids Research, 48(D1):D402–D406, 11 2019

    Charles J Norsigian, Neha Pusarla, John Luke McConn, James T Yurkovich, Andreas DrÃCger, Bernhard O Palsson, and Zachary King. Bigg models 2020: multi-strain genome-scale models and expansion across the phylogenetic tree.Nucleic Acids Research, 48(D1):D402–D406, 11 2019

    work page 2020

  5. [5]

    Metabolic modelling of microbes: the flux- balance approach.Environmental microbiology, 4(3), 2002

    Jeremy S Edwards, Markus Covert, and Bernhard Palsson. Metabolic modelling of microbes: the flux- balance approach.Environmental microbiology, 4(3), 2002

    work page 2002

  6. [6]

    Flux balance analysis of biological systems: applications and challenges.Briefings in bioinformatics, 10(4):435–449, 2009

    Karthik Raman and Nagasuma Chandra. Flux balance analysis of biological systems: applications and challenges.Briefings in bioinformatics, 10(4):435–449, 2009

    work page 2009

  7. [7]

    Metabolic fluxes and beyond–systems biology understanding and engineering of microbial metabolism.Applied microbiology and biotechnology, 88(5):1065–1075, 2010

    Michael Kohlstedt, Judith Becker, and Christoph Wittmann. Metabolic fluxes and beyond–systems biology understanding and engineering of microbial metabolism.Applied microbiology and biotechnology, 88(5):1065–1075, 2010

    work page 2010

  8. [8]

    Constraining the metabolic genotype– phenotype relationship using a phylogeny of in silico methods.Nature Reviews Microbiology, 10(4):291– 305, 2012

    Nathan E Lewis, Harish Nagarajan, and Bernhard O Palsson. Constraining the metabolic genotype– phenotype relationship using a phylogeny of in silico methods.Nature Reviews Microbiology, 10(4):291– 305, 2012

    work page 2012

  9. [9]

    Analysis of optimality in natural and perturbed metabolic networks.Proceedings of the national academy of sciences, 99(23):15112–15117, 2002

    Daniel Segre, Dennis Vitkup, and George M Church. Analysis of optimality in natural and perturbed metabolic networks.Proceedings of the national academy of sciences, 99(23):15112–15117, 2002

    work page 2002

  10. [10]

    Regulatory on/off minimization of metabolic flux changes after genetic perturbations.Proceedings of the national academy of sciences, 102(21):7695–7700, 2005

    Tomer Shlomi, Omer Berkman, and Eytan Ruppin. Regulatory on/off minimization of metabolic flux changes after genetic perturbations.Proceedings of the national academy of sciences, 102(21):7695–7700, 2005

    work page 2005

  11. [11]

    Genome-scale metabolic models applied to human health and disease.Wiley Interdisciplinary Reviews: Systems Biology and Medicine, 9(6):e1393, 2017

    Daniel J Cook and Jens Nielsen. Genome-scale metabolic models applied to human health and disease.Wiley Interdisciplinary Reviews: Systems Biology and Medicine, 9(6):e1393, 2017

    work page 2017

  12. [12]

    Context-specific metabolic networks are consistent with experiments.PLoS computational biology, 4(5):e1000082, 2008

    Scott A Becker and Bernhard O Palsson. Context-specific metabolic networks are consistent with experiments.PLoS computational biology, 4(5):e1000082, 2008

    work page 2008

  13. [13]

    Network-based prediction of human tissue-specific metabolism.Nature biotechnology, 26(9):1003–1010, 2008

    Tomer Shlomi, Moran N Cabili, Markus J Herrgård, Bernhard Ø Palsson, and Eytan Ruppin. Network-based prediction of human tissue-specific metabolism.Nature biotechnology, 26(9):1003–1010, 2008

    work page 2008

  14. [14]

    Computational reconstruction of tissue-specific metabolic models: application to human liver metabolism.Molecular systems biology, 6(1):401, 2010

    Livnat Jerby, Tomer Shlomi, and Eytan Ruppin. Computational reconstruction of tissue-specific metabolic models: application to human liver metabolism.Molecular systems biology, 6(1):401, 2010

    work page 2010

  15. [15]

    Reconstruction of genome-scale active metabolic networks for 69 human cell types and 16 cancer types using init.PLoS computational biology, 8(5):e1002518, 2012

    Rasmus Agren, Sergio Bordel, Adil Mardinoglu, Natapol Pornputtapong, Intawat Nookaew, and Jens Nielsen. Reconstruction of genome-scale active metabolic networks for 69 human cell types and 16 cancer types using init.PLoS computational biology, 8(5):e1002518, 2012

    work page 2012

  16. [16]

    Metabolic modeling of single th17 cells reveals regulators of autoimmunity.Cell, 184(16):4168–4185, 2021

    Allon Wagner, Chao Wang, Johannes Fessler, David DeTomaso, Julian Avila-Pacheco, James Kaminski, Sarah Zaghouani, Elena Christian, Pratiksha Thakore, Brandon Schellhaass, et al. Metabolic modeling of single th17 cells reveals regulators of autoimmunity.Cell, 184(16):4168–4185, 2021

    work page 2021

  17. [17]

    John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin ÅœÃdek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A. A. Kohl, Andrew J. Ballard, Andrew Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, Jonas Adler, Trevor Back, Stig Petersen, David Reima...

    work page 2021

  18. [18]

    Systems-level analysis of mechanisms regulating yeast metabolic flux.Science, 354(6311):aaf2786, 2016

    Sean R Hackett, Vito RT Zanotelli, Wenxin Xu, Jonathan Goya, Junyoung O Park, David H Perlman, Patrick A Gibney, David Botstein, John D Storey, and Joshua D Rabinowitz. Systems-level analysis of mechanisms regulating yeast metabolic flux.Science, 354(6311):aaf2786, 2016

    work page 2016

  19. [19]

    Cobrapy: constraints-based reconstruction and analysis for python.BMC systems biology, 7:1–6, 2013

    Ali Ebrahim, Joshua A Lerman, Bernhard O Palsson, and Daniel R Hyduke. Cobrapy: constraints-based reconstruction and analysis for python.BMC systems biology, 7:1–6, 2013

    work page 2013

  20. [20]

    Regulation of gene expression in flux balance models of metabolism.Journal of theoretical biology, 213(1):73–88, 2001

    Markus W Covert, Christophe H Schilling, and Bernhard Palsson. Regulation of gene expression in flux balance models of metabolism.Journal of theoretical biology, 213(1):73–88, 2001

    work page 2001

  21. [21]

    Accurate flux predictions using tissue-specific gene expression in plant metabolic modeling.Bioinformatics, 39(5):btad186, 2023

    Joshua AM Kaste and Yair Shachar-Hill. Accurate flux predictions using tissue-specific gene expression in plant metabolic modeling.Bioinformatics, 39(5):btad186, 2023

    work page 2023

  22. [22]

    Systematic evaluation of methods for integration of transcriptomic data into constraint-based models of metabolism.PLoS computational biology, 10(4):e1003580, 2014

    Daniel Machado and Markus Herrgård. Systematic evaluation of methods for integration of transcriptomic data into constraint-based models of metabolism.PLoS computational biology, 10(4):e1003580, 2014

    work page 2014

  23. [23]

    & Odell P. L. Piziak R. Affine projections.Computers and Mathematics with Applications, 2004

    work page 2004

  24. [24]

    Minimal reaction sets for escherichia coli metabolism under different growth requirements and uptake environments.Biotechnology progress, 17(5):791–797, 2001

    Anthony P Burgard, Shankar Vaidyaraman, and Costas D Maranas. Minimal reaction sets for escherichia coli metabolism under different growth requirements and uptake environments.Biotechnology progress, 17(5):791–797, 2001

    work page 2001

  25. [25]

    The effects of alternate optimal solutions in constraint-based genome-scale metabolic models.Metabolic engineering, 5(4):264–276, 2003

    Radhakrishnan Mahadevan and Chrisophe H Schilling. The effects of alternate optimal solutions in constraint-based genome-scale metabolic models.Metabolic engineering, 5(4):264–276, 2003

    work page 2003

  26. [26]

    M. Jain, R. Nilsson, S. Sharma, N. Madhusudhan, T. Kitami, A.L. Souza, and V .K. Mootha. Metabolite profiling identifies a key role for glycine in rapid cancer cell proliferation.Science, 336(6084):1040–1044, 2012

    work page 2012

  27. [27]

    Evaluation of 13c isotopic tracers for metabolic flux analysis in mammalian cells.Journal of biotechnology, 144(3):167–174, 2009

    Christian M Metallo, Jason L Walther, and Gregory Stephanopoulos. Evaluation of 13c isotopic tracers for metabolic flux analysis in mammalian cells.Journal of biotechnology, 144(3):167–174, 2009

    work page 2009

  28. [28]

    Estradiol stimulates the biosynthetic pathways of breast cancer cells: detection by metabolic flux analysis.Metabolic engineering, 8(6):639–652, 2006

    Neil S Forbes, Adam L Meadows, Douglas S Clark, and Harvey W Blanch. Estradiol stimulates the biosynthetic pathways of breast cancer cells: detection by metabolic flux analysis.Metabolic engineering, 8(6):639–652, 2006

    work page 2006

  29. [29]

    Zielinski, Neema Jamshidi, Amanda J

    Daniel C. Zielinski, Neema Jamshidi, Amanda J. Corbett, Aarash Bordbar, Allison Thomas, and Bernhard ø. Palsson. Systems biology analysis of drivers underlying hallmarks of cancer cell metabolism.Scientific Reports, 7(1):41241, 2017

    work page 2017

  30. [30]

    The cancer cell line encyclopedia enables predictive modelling of anticancer drug sensitivity.Nature, 483(7391):603–607, 2012

    Jordi Barretina, Giordano Caponigro, Nicolas Stransky, Kavitha Venkatesan, Adam A Margolin, Sungjoon Kim, Christopher J Wilson, Joseph Lehár, Gregory V Kryukov, Dmitriy Sonkin, et al. The cancer cell line encyclopedia enables predictive modelling of anticancer drug sensitivity.Nature, 483(7391):603–607, 2012

    work page 2012

  31. [31]

    The cancer genome atlast program (tcga).https://www.cancer.gov/tcga

    National Cancer Institute. The cancer genome atlast program (tcga).https://www.cancer.gov/tcga. Accessed: 2025-12-11

    work page 2025

  32. [32]

    Drake.Python 3 Reference Manual

    Guido Van Rossum and Fred L. Drake.Python 3 Reference Manual. CreateSpace, Scotts Valley, CA, 2009

    work page 2009

  33. [33]

    Troppo-a python framework for the reconstruction of context-specific metabolic models

    Jorge Ferreira, Vítor Vieira, Jorge Gomes, Sara Correia, and Miguel Rocha. Troppo-a python framework for the reconstruction of context-specific metabolic models. InPractical Applications of Computational Biology and Bioinformatics, 13th International Conference, pages 146–153. Springer, 2020

    work page 2020

  34. [34]

    Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, Stéfan J

    Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, Matt Haberland, Tyler Reddy, David Cournapeau, Evgeni Burovski, Pearu Peterson, Warren Weckesser, Jonathan Bright, Stéfan J. van der Walt, Matthew Brett, Joshua Wilson, K. Jarrod Millman, Nikolay Mayorov, Andrew R. J. Nelson, Eric Jones, Robert Kern, Eric Larson, C J Carey, ˙Ilhan Polat, Yu Feng, Eric W. M...

    work page 2020

  35. [35]

    Harris, K

    Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, Ralf Gommers, Pauli Virtanen, David Cournapeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J. Smith, Robert Kern, Matti Picus, Stephan Hoyer, Marten H. van Kerkwijk, Matthew Brett, Allan Haldane, Jaime Fernández del Río, Mark Wiebe, Pearu Peterson, Pierre Gérard-Marchant, Kevin Shepp...

    work page 2020

  36. [36]

    Pytorch: An imperative style, high-performance deep learning library.Advances in neural information processing systems, 32, 2019

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high-performance deep learning library.Advances in neural information processing systems, 32, 2019

    work page 2019

  37. [37]

    pandas-dev/pandas: Pandas, February 2020

    The pandas development team. pandas-dev/pandas: Pandas, February 2020

    work page 2020

  38. [38]

    Data Structures for Statistical Computing in Python

    Wes McKinney. Data Structures for Statistical Computing in Python. In Stéfan van der Walt and Jarrod Millman, editors,Proceedings of the 9th Python in Science Conference, pages 56 – 61, 2010

    work page 2010

  39. [39]

    Michael L. Waskom. seaborn: statistical data visualization.Journal of Open Source Software, 6(60):3021, 2021

    work page 2021

  40. [40]

    J. D. Hunter. Matplotlib: A 2d graphics environment.Computing in Science & Engineering, 9(3):90–95, 2007

    work page 2007

  41. [41]

    Jupyter notebooks - a publishing format for reproducible computational workflows

    Thomas Kluyver, Benjamin Ragan-Kelley, Fernando Pérez, Brian Granger, Matthias Bussonnier, Jonathan Frederic, Kyle Kelley, Jessica Hamrick, Jason Grout, Sylvain Corlay, Paul Ivanov, Damián Avila, Safia Abdalla, Carol Willing, and Jupyter development team. Jupyter notebooks - a publishing format for reproducible computational workflows. In Fernando Loizide...

    work page 2016

  42. [42]

    Context-specific genome-scale metabolic modelling and its application to the analysis of covid-19 metabolic signatures.Metabolites, 13(1), 2023

    Miha MoÅ¡kon and Tadeja ReÅŸen. Context-specific genome-scale metabolic modelling and its application to the analysis of covid-19 metabolic signatures.Metabolites, 13(1), 2023. Appendix A Background A.1 Constraint-Based Metabolic Modeling and Flux Balance Analysis LetS∈R m×r be a stoichiometric matrix whose rows correspond to metabolites and columns to re...

    work page 2023